Commission, May 7, 2018, No M.8444

COMMISSION DECISION of 7.5.2018

declaring a concentration to be compatible with the internal market and the EEA agreement

(Case M.8444 – ArcelorMittal/Ilva)

(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 undertakings (1), and in particular Article 8(2) thereof,

Having regard to the Commission's decision of 8 November 2017 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 (2), Having regard to the final report of the Hearing Officer in this case (3), Whereas:

(1)     On 21 September 2017 the Commission received a notification of a proposed concentration pursuant to Article 4 of Regulation (EC) No 139/2004 (the ‘Merger Regulation’) by which ArcelorMittal S.A. (‘ArcelorMittal’, ‘AM’, or the ‘Notifying Party’) acquires sole control of certain assets of the Ilva Group: namely the business of Ilva S.p.A and a number of its subsidiaries, namely Ilva S.p.A, Taranto Energia S.r.l, Ilva Servizi Marittimi S.p.A, Tillet S.a.s and Socova S.a.s. (‘the Transaction’). (4)

1.  THE PARTIES

(2)     ArcelorMittal S.A (‘ArcelorMittal’) is based in Luxembourg and is the largest steel producer in the world. It is a multinational steel manufacturing and mining company, whose principal business is the production, distribution, marketing and sale of steel products for various applications, including automotive, construction, household appliances and packaging.

(3)     The Ilva Group, which includes Ilva S.p.A. (‘Ilva’), is based in Italy. One of Europe's largest steelmakers, it is active in the production, processing and distribution of carbon steel. Ilva operates steel plants in Italy (including plants in Taranto, Genoa, and Novi Ligure) and various steel service centres (‘SSCs’) for product distribution.

1.1. The Ilva assets

(4)     The Transaction does not involve the acquisition of the entire Ilva Group, but is limited to certain business units which belong to the following entities of the Ilva Group: (1) Ilva S.p.A. and a number of its subsidiaries, which are all in extraordinary administration  (Amministrazione  Straordinaria),  namely  –  (2)   Ilvaform     S.p.A;

(3) Taranto  energia  S.r.l;  (4)  Ilva  Servizi  Marittimi  S.p.A;  (5)  Tillet  S.a.S.; and

(6) Socova S.a.S. Those business units consist most notably of Ilva's plants in Taranto, Genova and Novi Ligure, its steel service centres in Salerno (Italy) and Sénas (France), Ilva's maritime fleet, the power station operating the Taranto plant and certain distribution outlets in Italy and France.

(5)     According to the information provided by the Notifying Party, the  Transaction  covers only the assets of those companies and there would be no legal succession of those entities. The assets of the Ilva Group that are subject to the Transaction are hereinafter referred to as ‘the Ilva assets’, while the Ilva Group as a competitor of ArcelorMittal prior to the transaction will be referred to as ‘Ilva’.

(6)     Figure 1 shows the structure of the Ilva Group which is currently under extraordinary administration and the Ilva assets within the perimeter of the Transaction.

Figure 1 – [Internal document] (5)

[…]

(7)     For the purpose of this Decision, ArcelorMittal and the Ilva are hereinafter jointly referred to as ‘the Parties’ while ArcelorMittal is referred to as ‘the Notifying Party’.

2.  THE OPERATION AND THE CONCENTRATION

(8)     Ilva Group has been owned by the Riva Group since 1996 and was previously controlled by the Italian State through IRI – Istituto per la Ricostruzione Industriale. It  entered  into  insolvency  proceedings  (Amministrazione   Straordinaria)   in March 2015. Since then, Ilva has been run by three government-appointed extraordinary commissioners.

(9)     In January 2016, the Italian Government, as part of the special insolvency proceedings, launched a competitive tender for the sale of the Ilva assets. ArcelorMittal submitted a binding offer to acquire the Ilva assets on 6 March 2017 through AM InvestCo Italy S.r.l. (‘AM Consortium’), a solely-controlled subsidiary of  ArcelorMittal,  of  which  ArcelorMittal  currently  owns  […]%. (6) The remaining […]% is held by Marcegaglia Carbon Steel S.p.A (‘Marcegaglia’), an Italian steel company which is both a competitor of and one of the largest customers of Ilva. […], at the closing of the Transaction, ArcelorMittal would own approximately […]% of AM Consortium, while Marcegaglia and Intesa Sanpaolo S.p.A. (an Italian bank and one of Ilva Group’s creditors) would each own small shareholdings of circa […]%, neither of which will confer joint control over the Ilva assets. (7)

(10)    As a result of the Transaction, ArcelorMittal would have sole control of the Ilva assets.

(11)    The Transaction constitutes a concentration within the meaning of Article 3(1)(b) of the Merger Regulation.

3. UNION DIMENSION

(12)    The Transaction has a Union dimension as the turnover  thresholds  set  out  in  Article 1(2) of the Merger Regulation are met. The combined aggregate worldwide turnover of the Parties is more    than EUR 5 000 million (ArcelorMittal EUR 51 306 million, the Ilva assets EUR […] million) (8) and the aggregate Union-wide turnover of each of the Parties is more than EUR 250 million (ArcelorMittal EUR […] million, the Ilva assets EUR […] million). Neither ArcelorMittal nor the Ilva assets achieve more than two-thirds of their Union-wide turnover within one and the same Union Member State.

(13)    The notified operation therefore has a Union dimension.

4. THE PROCEDURE

4.1. The administrative procedure

(14)    During the Phase I investigation, the Commission contacted a number of market participants (including customers and competitors of the Parties) and requested information both through seven electronic questionnaires pursuant to Article 11 of  the Merger Regulation and telephone calls.

(15)    On 12 October 2016, the Commission informed the Parties of the concerns resulting from the preliminary assessment of the Transaction during a 'State of Play' meeting.

(16)    On 19 October 2017, the Parties submitted commitments that included the divestment of [Parties' submissions]. Those commitments were not market tested.

(17)    Based on the results of the Phase I market investigation, the Commission found that the Transaction raised serious doubts as to its compatibility with the internal market and adopted a decision to initiate proceedings pursuant to Article 6(1)(c) of the Merger Regulation on 8 November 2017 (the 'Article 6(1)(c) decision').

(18)    On 10 November 2017, the Commission provided a number of key documents illustrating the nature of the case to the Notifying Party.

(19)    On 18 November 2017 ArcelorMittal submitted to the Commission its comments in response to the Article 6(1)(c) decision, a revised version of which was submitted on 19 November 2017.

(20)    On 22 November 2017, at a state of play meeting, the Commission provided the Parties with the opportunity to discuss orally the main issues raised in the response to the Article 6(1)(c) decision and indicated the matters on which it planned to focus its further investigative efforts.

(21)    During the Phase II market investigation, the Commission sent several requests for information to the Parties, as well as to third parties. The Commission held several calls with market participants, and also sent requests for information in the form of three questionnaires, in addition to the seven questionnaires sent out before the initiation of the proceedings, to a total of more than 1800 addressees. Furthermore, the Commission analysed internal documents of the Parties.

(22)    On 15 December 2017, in agreement with the Notifying Party, the Commission decided to extend the time period set out in the first subparagraphs of Article 10(3) of the Merger Regulation by a total of five working days in accordance with the third sentence of the second subparagraph of Article 10(3) of the Merger Regulation.

(23)    Following the results of the Phase II market investigation, a state of play meeting  was held on 16 January 2017, in order to inform the Notifying Party of the preliminary results of the Phase II market investigation and the scope of the preliminary concerns on which the Commission planned to issue a Statement of Objections.

(24)    On 18 January 2018, the Commission adopted a Statement of Objections (‘SO’), which was sent to the Notifying Party on the same day. According to the SO, the Commission came to the preliminary view that the Transaction would significantly impede effective competition in a substantial part of the internal market within the meaning of Article 2 of the Merger Regulation due to (i) horizontal non-coordinated effects in the markets for the production and supply of hot rolled flat carbon steel, cold rolled steel, hot dip galvanised steel and electro galvanised steel products in    the EEA; and (ii) horizontal coordinated effects in a number of markets for flat carbon steel in the EEA. 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.

(25)    The Notifying Party was granted access to the file on 19 January 2018 via CD-ROM. A data room was organised from 19 January to 29 January 2018 allowing the economic advisors of ArcelorMittal to verify confidential information of a quantitative nature, which formed part of the Commission's file. Further documents were  sent  on  24  January, 1 February, 2 March, 26 March, 28 March  and 26 April 2018. A second data room was organised from 1 March to 2 March 2018.

(26)    The Ilva Group submitted its observations to the SO on 1 February 2018.

(27)    ArcelorMittal was given until 2 February 2018 to reply to the SO and eventually submitted its reply on the morning of 3 February 2018 (the 'Reply to the SO').

(28)    Both ArcelorMittal and the Ilva Group requested to be heard orally.

(29)    ThyssenKrupp AG, Tata Steel Limited and Marcegaglia made applications to the Hearing Officer to be admitted as interested third persons in the proceedings.

(30)    All interested third persons were provided with a non-confidential version of the SO and given a time-limit within which to submit their responses. Given the urgency of the  proceedings,  ThyssenKrupp  AG,  Tata  Steel  Limited  and  Marcegaglia   were allowed to make known their views at the oral hearing prior to the submission of  their  written   comments   pursuant   to   Article   16(2)   of   Commission  Regulation (EC) 802/2004 (9).

(31)    An oral hearing was held on 8 February 2018.

(32)    On 22 February 2018, in accordance with the third sentence of the second subparagraph of Article 10(3) of the Merger Regulation and in agreement with the Notifying Party, the Commission decided to extend the time period set out in the first subparagraph of Article 10(3) of the Merger Regulation by a total of 11 working days. This decision was taken to allow the Commission and the Parties to have sufficient time to discuss and thoroughly assess any remedy proposal that may be submitted by the Notifying Party. Accordingly, the deadline for a Commission decision in this proceeding was extended until 19 April 2018.

(33)    A Letter of Facts – evidence corroborating the objections set out in the SO – was sent to the Notifying Party on 28 February 2018. The Notifying Party submitted its comments on the Letter of Facts on 9 March 2018 ('Reply to the Letter of Facts').

(34)    The Notifying Party was then granted further access to the file on 2 March 2018 via CD-ROM. A data room was organised from 1 March to 2 March 2018.

(35)    On 5 March and 11 March 2018, the Notifying Party submitted draft commitments in order to address the competition concerns identified in the SO. Consequently, on 12 March 2018, the period for the adoption of a final Decision was extended by 4 working days pursuant to Article 10(3) second subparagraph, third sentence of the Merger Regulation to allow for the Commission and the Parties to have sufficient time to discuss and thoroughly assess any formal remedy proposal that may be submitted by the Notifying Party. Accordingly on 12 March 2018, the deadline for a Commission decision in this proceeding was extended until 25 April 2018.

(36)    On 15 March 2018, 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 15 March 2018').

(37)    On 15 March 2018, the Commission launched a market test of the Commitments      of 15 March 2018.

(38)    The Notifying Party was then granted further access to the file on 28 March 2018 via email.

(39)    On 11 April 2018,  the  Notifying  Party  submitted  a  final  set  of  commitments  (the 'Final Commitments').

(40)    The meeting of the Advisory Committee took place on 2 May 2018.

(41)    The Hearing Officer issued his final report on 3 May 2018.

4.2. Considerations regarding the reliability of the replies to the Commission's market investigation

4.2.1.Introduction

(42)    As outlined in Section 4.2.3, in the course of its investigation, the Commission found evidence that the Notifying Party had reached out to certain customers of Ilva to discuss their responses to the Commission's investigation. The Commission set out this evidence in the SO and noted that while it may further investigate the issue, it would take account of the fact that such interactions between the Notifying Party and its customers may affect the reliability of the customers' replies to the questionnaires.

4.2.2. The Notifying Party's views

(43)    According to the Notifying Party, the SO points to a limited set of documents that describe ArcelorMittal's outreach to inform key customers about the Transaction, answer questions that customers might have about the implications of the  Transaction for their business and explain the benefits ArcelorMittal considers the Transaction would bring to their customers. In this regard, the Notifying Party submits that the following points bear mention:

(1)   According to the Notifying Party, customer outreach is a normal and legitimate business practice during significant M&A activity.

(2)   According to the Notifying Party, there is no evidence that ArcelorMittal sought to influence responses to the Commission’s market investigation. Rather, according to the Notifying Party, the documents cited show a  legitimate and normal customer communications plan, and a ‘very limited’ exercise in canvassing feedback from a small number of Italian customers where three topics were discussed: [internal document]. According to the Notifying Party that limited exercise was informative in nature and there is no evidence of any attempt to influence those customers future replies (either explicit or implicit).

(3)  Further, according to the Notifying Party, the responses to the Commission’s market investigation confirm that ArcelorMittal did not seek to influence responses.

(4)   Finally, the Notifying Party submits that there is no basis in statute or  precedent to warrant further ‘investigation’ on this issue, as the SO suggests and none of the issues summarised above justify a departure from the Commission's obligation to take all relevant evidence into account, including customer support for the Transaction when reaching a decision under the Merger Regulation. In this regard, the Notifying Party submits that the results of the Commission's market investigation were markedly more supportive of the Transaction than the SO suggests and that it is incumbent on the Commission to take into account all relevant evidence in an even handed fashion.(10)

4.2.3. The Commission's assessment

(44)    The  Commission  takes  due  note  of  the  arguments  of  the  Notifying  Party     (see Section 4.2.2) and the case law quoted by the Notifying Party in its response to the SO. In this regard, the Commission notes that in ABB Asea Brown Boveri v Commission, (11) the Court held that ‘[t]he guarantees conferred by the Community legal order in administrative proceedings include, in particular, the duty of the competent institution to examine carefully and impartially all the relevant aspects of the individual case.’ In this regard, the Commission also notes that in that case the Court rejected the plea alleging infringement of the principle of sound administration given, inter alia, that the allegations in question were in fact supported by the evidence gathered by the Commission.

(45)    As regards the Commission's investigation in this case, as outlined in the SO, the Commission found documentary evidence that the Notifying Party initiated contacts and held discussions with certain customers of the Ilva Group concerning their responses to the Commission's investigation, as further detailed in the following paragraphs.

(46)    As acknowledged by the Notifying Party, ArcelorMittal pursued a customer communication plan to discuss the Transaction, which included customer events, personal visits, and letters. (12)  In this framework, […]. (13)

(47)    In a summary document, […]. (14)

(48)    This behaviour is also demonstrated by […]: […] (15)

(49)    On the basis of the evidence quoted in this Section, the Commission considers that the intention of such exchanges went beyond ArcelorMittal's ‘interest in communicating and explaining to individual customers what it considers to be the benefits of the Transaction for customers, to answer questions, and dispel possible misunderstandings.’ (16) Rather, as indicated more particularly in the email cited in recital (48), there was clearly an intention to influence the outcome of the market investigation in favour of the Transaction. The exchanges with customers concerned [internal document]. Considering its obligation ‘to examine carefully and impartially all the relevant aspects of the individual case’ (17), the Commission finds that these interactions likely affect the reliability of replies to questionnaires in the  second phase of the investigation, particularly where they concern opinions about the Transaction as opposed to the transmission of factual information. The Commission has taken this into account when assessing individual replies to the market investigation.

5. INTRODUCTION TO THE FLAT CARBON STEEL INDUSTRY

5.1. Value chain and production process of flat carbon steel

5.1.1. Production process of flat carbon steel

(50)    Carbon steel is based on iron and carbon and it contains little or no alloying elements. (18) Carbon steel production consists of two main stages: (i) the production of crude steel and semi-finished products, and (ii) the production of final products.

(51)    There are two principal processes for the production of crude steel and semi-finished products: (i) the  so-called  integrated  or  blast  furnace  (‘BF’)/basic  oxygen  furnace (‘BOF’) route and (ii) the electric arc furnace (‘EAF’) route. Both methods involve the production or use of an iron-containing material – liquid hot metal (liquid carbon-saturated  iron),  pig  iron  (solid  carbon-saturated  iron),  direct   reduced iron (‘DRI’) or scrap.

(52)    The integrated route involves the production of liquid hot metal from a mixture of iron ore, coke and limestone in a blast furnace. The liquid hot metal, which has a  high carbon content and would thus be brittle in its solid state (pig iron) is refined into steel in a basic oxygen furnace (‘BOF’) where also scrap may be added. (19) In the BOF process, the carbon content of the liquid metal is reduced by blowing oxygen into it; the oxygen combines with carbon to form gaseous compounds that leave the liquid thereby reducing its carbon content. During the BOF process, or in a separate secondary steelmaking, the composition of the steel is adjusted to give it the desired qualities. The adjustment may include the use of appropriate quantities of various alloying elements.

(53)    The EAF route involves melting of scrap into liquid steel in a special furnace, sometimes together with pig iron or DRI. During this process, or in a separate secondary steelmaking, the composition of the steel is adjusted to give it the desired qualities. The adjustment may include the use of appropriate quantities of various alloying elements.

(54)    Once the liquid steel has been produced – be it through the integrated or EAF route,  it is cast and cooled in continuous casting machines to produce semi-finished carbon steel products. There are three main types of semi-finished carbon steel: slabs, blooms and billets. Slabs are used in the production of finished flat products while blooms and billets are used in the production of finished long products. As the Transaction only concerns flat products, the production process of long products is not discussed further in this Decision.

(55)    To produce finished flat carbon steel products, the slabs are cooled and transported to a hot rolling mill either on-site or at another location for further processing into thinner, usable plates or strips. In the hot rolling mill the slabs are reheated to a desired temperature in a reheat furnace and subsequently rolled into different thicknesses through a series of stands. There are two main types of hot rolling mills: plate mills that produce quarto plates in a special four stand mill and strip mills that produce hot rolled strips that are thinner than quarto plates and that can be coiled  after rolling.

(56)    If required, hot rolled coils can be further rolled down to thinner gauges (and better surface quality) in cold rolling mills. Finally, the cold rolled (and sometimes hot rolled) coils can also be coated with thin layers of metals or organic materials, resulting for instance in galvanised or organic coated products, among others.

(57)    The production process for flat carbon steel products is illustrated in Figure 2.

5.1.2.  Description of flat carbon steel products and their distribution

5.1.2.1. Semi-finished products

(58)    The semi-finished carbon steel products used as input in the finished flat product production are slabs. Slabs are typically produced from the hot liquid steel at the  steel shop through continuous casting. Continuous casting starts as the molten metal is transported from the basic oxygen furnace or an electric arc furnace in a ladle and poured into a tundish which provides for a reservoir of liquid steel for the casting process. In the casting, the liquid steel is continuously cast through a mould to form the desired width and thickness of the slab, which is eventually cut into usable lengths at the  end  of  the  process.  The  continuous  casting  process  is  illustrated in Figure 3.

5.1.2.2. Quarto plates

(59)    Quarto plates (‘QP’) are non-coiled hot rolled flat carbon steel products that are produced in dedicated plate mills after reheating the slabs to the desired temperature in reheat furnaces. QP differ from other hot rolled flat carbon steel products in their dimensions, being in particular thicker than strip products. They are used in applications that call for thick steel, such as at shipyards, boiler-making, nuclear and the oil & gas industry.

5.1.2.3.  Hot rolled products (excluding QP)

(60)    Hot rolled flat carbon steel products other than quarto plates (‘HR’) are strip products that are produced from slabs through hot rolling in a strip mill after reheating the slabs to the desired temperature in reheat furnaces. HR products differ from QP in that they are thinner and are typically coiled at the end of the hot rolling process. HR is also the input for downstream products discussed in this Decision, including cold rolled, hot dip galvanised, electro-galvanised, organic coated and metallic coated  steel for packaging.

(61)    Figure 4 shows a slab exiting a reheat furnace, ready to enter the rolling phase in a hot strip mill.

(62)    Figure 5 shows a finished HR strip being coiled at the end of the hot rolling mill.

5.1.2.4. Cold rolled products

(63)    Cold rolled flat carbon steel products (‘CR’) are the result of the further rolling of  HR in cold rolling mills. Cold rolling affects the basic properties of the product by reducing thickness, improving dimensional consistency and providing a smoother surface.

(64)    CR is commonly used as an input in the production of coated steel products but may also be sold without further treatment into various applications, including construction, furniture manufacturing, welded tube making as well as packaging and machinery production.

(65)    A CR coil is shown in Figure 6.

5.1.2.5.  Galvanised products

(66)    Galvanised steel products (‘GS’) are flat carbon steel products that have been galvanised to improve their resistance to corrosion. Galvanisation involves coating the flat carbon steel strip with zinc, or a combination of elements with zinc being the primary one. GS is used in applications such as the automotive industry, construction and various engineering industry applications where superior corrosion resistance is required. Both HR and CR can be used as inputs for the production of galvanised products but the clear majority of GS produced by the Parties is made from CR (ArcelorMittal […]%; Ilva […]%).

(67)    GS may be obtained via two main production processes: (i) hot dip galvanisation and (ii) electro galvanisation. Hot dip galvanised products (‘HDG’) are produced through uncoiling and reheating the steel strip and feeding it through a bath of molten metal composed of zinc or zinc and other elements at an appropriate temperature. Electro galvanised products (‘EG’) are produced through the application of an electrolytic coating process which results in a zinc-containing coating on one or both sides of the strip.

(68)    GS coils are shown in Figure 7.

5.1.2.6. Organic coated products

(69)    Organic coated flat carbon steel products (‘OC’) are obtained by adding a  thin organic coating, often paint, on GS (or sometimes CR). Organic coating provides for a protective layer and an attractive physical appearance. OC is used in applications such as in construction or the production of furniture and white goods.

5.1.2.7. Metallic coated steel for packaging

(70)    Metallic coated steel for packaging consists of thin CR coils or sheets that have been coated with a fine layer of another metal, primarily tin or chromium, resulting in either tinplate (‘TP’) or electrolytic chromium coated steel (‘ECCS’). TP and ECCS are primarily used in the food and beverage industry to produce protective packages (such as food cans).

(71)    TP coils are shown in Figure 8.

5.1.2.8. Welded carbon steel tubes

(72)    Carbon steel tubes can be produced from flat and long steel products. The production of seamless carbon steel tubes requires input of long steel products, whereas welded carbon steel tubes are manufactured through further processing of flat steel products. Flat carbon steel products such as HR, CR and GS are the main input in the downstream production of welded carbon steel tubes. HR accounts  for  approximately […]% of input for the production of welded carbon steel tubes, while CR and GS account for the remainder.

(73)    Large diameter tubes can be distinguished from tubes with a smaller diameter, due to the differences in their production processes and fields of application. Large tubes have a diameter of more than 20 inches (508mm) for welded tubes and 24 inches  (610 mm) for seamless tubes.

5.1.2.9.  Distribution of flat carbon steel products

(74)    While steel mills tend to supply large orders of standard dimensions with longer lead times, distribution centres typically supply smaller lot sizes and have shorter lead times. In addition to ex-mill sales, flat carbon steel products are sold in the EEA through three types of distribution channels.

(75)    The three types of distribution  channels  are  the  following:  (i)  stockholding  centres ('SCs'); (ii) steel service centres ('SSCs'); and (iii) oxy-cutting centres. They each perform slightly different functions.

(76)    SCs perform a conventional wholesaling function of buying steel in bulk from steel producers and reselling it in smaller quantities. As their main activity, SCs supply products in standard dimensions without further processing.

(77)    SSCs purchase strip mill products from steel producers and cut the material based on their customers' requirements. Most SSCs process flat carbon steel products, but not QP or TP. TP is usually processed in dedicated SSCs.

(78)    Oxy-cutting centres process and distribute primarily QP.

5.1.3.   Primary steelmaking as the driver for the entire flat carbon steel value chain

5.1.3.1.  Dynamics of the steel value chain

(79)    The market position of flat carbon steel producers in the EEA is by and large determined by their primary steelmaking capacity (that is the capacity to produce liquid steel and slabs). As explained in more detail below, primary steelmaking is characterised on the one hand by high barriers to entry and expansion and, on the other hand, by limited flexibility to efficiently reduce production levels. Likewise,  the overall hot rolled capacity typically reflects the primary steelmaking  capacity and, in the EEA, is the closest proxy for the upstream market position determining a producer's strength on the markets for finished flat carbon steel products.

(80)    First, flat carbon steel production in the EEA is primarily based on the integrated route, that is to say the production of liquid steel through a combination of a blast furnace and a basic oxygen furnace. The EAF-route is relatively insignificant for the production of flat carbon steel products in the EEA in capacity terms and most producers have no EEA flat carbon steel EAF capacity at all. This has been acknowledged by the Notifying Party, which has submitted that approximately […]% of  flat  carbon  steel  in   the  EEA  originates  in  the  integrated  route while the remaining […]% is of scrap-based EAFs. (20) This applies to the Parties as well: ArcelorMittal produces predominantly through the integrated route with very limited EAF capacity for flat carbon steel products, and Ilva only produces through the integrated route.

(81)    Second, the supply-side dynamics of primary steelmaking in the EEA is driven by  the characteristics of the integrated route, which requires significant capital investments and which to a great extent relies on the simultaneous and constant operation of a blast furnace and a basic oxygen furnace, and often also a coking and a sintering plant.

(82)    In the first place, high capital expenditure and environmental regulation act as an effective barrier to entry or expansion. The Notifying Party estimates that the cost of greenfield investment into an integrated steel plant amounts […]. The Commission understands that the rolling mills account for only a portion of this cost, and the bulk of investment is associated with primary steelmaking capacity. This, compounded with modest demand growth in the EEA, is in line with the finding that, in recent years, the EEA saw no creation of new capacity for the integrated route technology.

(83)    In the second place, the integrated route has limited flexibility when it comes to adapting production volumes to demand fluctuations: for technical and economic reasons blast furnaces need to be run at, or close to, maximum capacity. The Notifying Party submits in this respect that production in a blast furnace can be decreased at best down to around […]% of full capacity. Restarting a blast furnace that has been blown down (that is where steel production has stopped) can take several days or weeks, entails high one off costs, and becomes increasingly difficult the longer the blast furnace has been idled. Where a blast furnace is long term idled, or mothballed, the cost of restarting production is estimated by the Notifying Party to around […] of capacity. (21)

(84)    Therefore, a steel manufacturer that bases its production on the integrated route is faced with limitations in its ability to alter the production volumes of liquid crude steel and – since casting typically takes place immediately following the production of hot liquid steel – semi-finished products, notably slabs. To allow for a viable operation, the producer needs to produce a certain volume with a blast furnace over extended periods of time, or not to produce with that blast furnace at all. Where supply outstrips demand, the supplier has thus few options to limit the output of blast furnaces in operation. The supplier can only decide to idle one or more blast furnaces altogether in order to align production and demand. On the other hand, the  production process for finished products (strips) is not subject to similarly strict limitations from a technical point of view and can be adjusted more flexibly. For example, the cost of restarting a mothballed strip mill / finishing asset is estimated by the Notifying Party to be EUR […] (22), which represents […]% or less of the costs for restarting a mothballed blast furnace.

(85)    Third, as a consequence of barriers to entry and expansion, as well as of inflexibility in scaling down production volumes, primary steelmaking capacity typically determines the overall available capacity for the entire flat carbon steel value chain. That  said,  hot  strip mills  are  not  necessarily integrated  with  primary steelmaking facilities, and they can process slabs sourced from another site of the steel producer  or sourced from third parties. This production model is occasionally observed in practice – for instance, […]. Compared to vertically integrated steel mills or intra- company slab supply, production based on slabs procured on the merchant market appears to be of much more limited scope and potentially less efficient. It may be employed as a temporary measure to bridge the imbalances between upstream primary steelmaking output and the requirements for the production of downstream strip products. (23)  The vast majority of the hot strip mill capacity (‘HR capacity’) in  the EEA mimics the primary steelmaking capacity (‘capacity for crude steel slabs’) located at the same or nearby site.

(86)    Fourth, given the interdependence of primary steelmaking and hot strip mills, the competitive position of integrated flat carbon steel producers (that is other than non- integrated re-rollers (24)) is driven by their HR capacity. From the upstream perspective of primary steelmaking, HR is the direct output of processing slabs. From the perspective of the production and supply of flat carbon steel, HR is both a final product and the input for all other finished flat products (including CR, HDG, EG, OC and steel for packaging). Thus, the competitiveness and the capacity for HR is on the one hand determined by primary steelmaking, and, on the other hand, also determines the conditions for the production and supply of downstream products.

(87)    Fifth, HR can either be sold on the merchant market or further processed into downstream finished products by the integrated producer. An integrated producer active on the various levels of the flat carbon steel value chain is thus faced with a choice as to at which level of the production chain it sells its steel. Typically the value of steel increases the further it is processed and, moreover, a producer that has installed capacity for the downstream products and needs to pay for the fixed costs so incurred may, in general, be incentivised to employ the existing capacity to produce products on the downstream market. Therefore, an integrated steel manufacturer may be incentivised – to the extent possible in the prevalent market conditions – to direct its HR products into further processing internally. The steel kept by the integrated steel producer for further processing in its own processing facilities is part of the so- called ‘captive market’ of steel, as opposed to the 'merchant market', where the steel is sold by the integrated producers to other steel manufacturers (notably re-rollers), distributors and industrial customers.

(88)    However, steel producers do not reserve a given primary/HR capacity solely to the production destined for the merchant market, or for captive use, but use their plants  to feed both channels. Given the aforementioned lack of flexibility in adapting the output to demand fluctuations, conditions on the merchant market will not only influence the volume to be sold to third parties, but also HR for captive use. The producer may at least partly reallocate capacity to the channel that provides a more attractive return. For example, if the prices on the merchant market for HR drop, the supplier may reallocate more capacity to captive production for downstream products with a higher added value. The same may occur where there is a shortage on a downstream market, for example for HDG. In other words, the suppliers may leverage their production available to the merchant market in order to react to developments   affecting   the   downstream   products,   and   vice   versa. Moreover, suppliers may align their pricing policy for merchant sales of HR and for the sales of downstream products based on HR (for example, by virtue of the base price plus extras pricing model (25)).

(89)    Consequently, the higher the overall HR production, the stronger the supplier’s pricing power both directly in the market for HR and in the related downstream markets. Therefore, the competitive assessment should not isolate suppliers’ position in the merchant channel from the remainder of their HR production and capacity.

(90)    Finally, non-integrated producers (‘re-rollers’) that source HR coils and then further process them into downstream products such as CR and GS are typical customers of HR coils on the merchant market, and they compete in the downstream CR and GS markets with the integrated producers. Nonetheless, re-rollers depend on the supply  of the input HR coils on the merchant market by the EEA integrated producers (and/or imports) and, hence, their market position is affected by the availability and conditions of the supply of HR on the merchant market by the integrated suppliers (and/or imports). The most prominent re-roller in the EEA is Marcegaglia in Italy, […].

(91)    The feedback received from market participants during the Commission's market investigation indicates that the competitiveness of re-rollers may be negatively affected by their dependence on sourcing HR coils from EEA integrated producers and/or imports:

'Not controlling the primary steel production increases the risk profile of the business.' (26)

'An integrated supplier can design the steel almost tailor made for the energy industry since specifications are to some extent unique for each customer and application of energy pipes. So it would be almost impossible for a non-integrated steel producer to achieve these qualities.' (27)

'Manufacturers who do not have the entire production process often acquire the basic products for the processing of their products from very different sources and sometimes they bring discontinuous qualitative results on the finished product. So in principle a supplier is preferable which has the entire production process.' (28)

'Marcegaglia (the most important re-roller / re-roller) has quality generally lower than steel mills; range of products generally lower than steel mills; containing of products over time much lower than steel mills. [Our company] does not know the reinsurers' business strategies but believes that re-rollers generally buy HR coils where these have favourable economic conditions. Therefore the HR suppliers change and therefore the characteristics of the materials also change downstream (CR and GS) that are produced from the HR. However, this aspect does not occur with integrated steel mills: these have their own "recipes" of production and tend to keep them stable in the time.' (29)

(92)    The potential exposure of re-rollers to shocks, and particularly measures on imports, is illustrated in […]: (30)

[…].

5.1.3.2.      Example: Development of the Ilva Group’s production

(93)    Preference for the captive use of HR may be observed, for instance in the development of the Ilva Group’s production following reductions in its hot metal and HR production. While  the  Ilva  Group’s  production  of  HR  decreased  from  […] in 2012 to […] in 2016, that is by […]%, its CR production only decreased from […] to […] ([…]%) and its GS production actually increased from […] to […]. This shows that Ilva Group favoured captive use of its HR production and consequently reduced its sales on the merchant market. The development of the Ilva Group’s production during 2012–2016 is shown in Table 1.

5.2. Introduction to the Parties' activities

5.2.1. ArcelorMittal

(94)    ArcelorMittal, (31) based in Luxembourg, is the leading flat carbon steelmaker in the EEA. It was formed in 2006 through the takeover and merger of West European steel maker Arcelor (Spain, France and Luxembourg) by Indian-owned Mittal Steel. ArcelorMittal produces approximately 10% of the world's steel. It is listed on the stock exchanges (32) of New  York  (MT),  Amsterdam  (MT),  Paris (MT),  Luxembourg (MT) and on the Spanish stock exchanges of Barcelona, Bilbao, Madrid and Valencia (MTS), and is a member of more than 120 indices. In 2016 ArcelorMittal   had   total   revenue   of   EUR   53   951   million   and    reached EUR 3 952 million EBITDA result.

(95)    ArcelorMittal is the only EEA manufacturer with an EEA-wide production network, with production facilities throughout the continent, including primary steel making facilities   in   Spain   (Sestao   and   Aviles),   France  (Fos-sur-Mer,   Atlantique and Florange), Belgium (Ghent), Germany (Bremen and Eisenhüttenstadt), Poland (Katowice Steelworks and Tadeusz Sendzimir Steelworks), the Czech Republic (Ostrava) and Romania (Hunedoara Steel Works and Galati Steel Works).

(96)    On top of its direct business, ArcelorMittal jointly controls with other partners a number of other manufacturing, trading and distribution activities, and holds  minority shareholdings in a number of other companies active in the steel manufacturing and supply chain.

(97)    In particular, with particular relevance to the EEA, ArcelorMittal has a 50% stake in a joint venture with Macsteel (steel trading and distribution) in the Netherlands; a 50% stake in a joint venture with Tamec (energy production and supply) in Poland and 45.33% stake in a joint venture with Borcelik (manufacturing and sale of steel), Turkey (33). Furthermore, ArcelorMittal has 33.43% ownership and voting rights in DHS Groups, Germany (steel manufacturing), 35% ownership in Gonvarri Steel Industries, Spain (steel manufacturing), 35% ownership and voting rights in  Gestamp, Spain (manufacturing of metal components); 28.47% ownership and voting rights in Stalprodukt S.A., Poland (production and distribution of steel products). ArcelorMittal also holds shares in the Turkish crude steel producer, 'Erdemir' (34) (Ereğli Demir ve Çelik Fabrikaları, Ereğli Iron and Steel Factories,), which has, according to the Notifying Party, around […] of crude steel capacity.

(98)    ArcelorMittal also has a wide distribution network. For instance, in Italy it has a 49% controlling interest in ArcelorMittal CLN Distribuzione Italia. In France, ArcelorMittal operates several SSCs through its  subsidiary  ArcelorMittal Distribution Solutions France.

(99)    Figure 9 shows the flat carbon steel production plants of ArcelorMittal in the EEA. ArcelorMittal's primary production assets are in Northern Continental Europe, in Eastern Europe and in Spain and South of France.

(100)    ArcelorMittal's primary steel making production site closest to Ilva is located in Southern France (Fos-sur-mer), producing both crude steel slabs and HR, but ArcelorMittal also operates further crude steel and HR production sites in Southern Europe, namely in Aviles in Northern Spain (and an EAF facility in Sestao, also in Northern Spain). In the South of Europe, ArcelorMittal also has three satellite plants in Italy (Avellino, Canossa and Piombino), and five downstream plants in Spain (Sagunto, Extebarri, Lesaka, Sestao, Gijon), which are active in the production of downstream flat carbon steel products, such as CR and HDG in Italy and CR, HDG, EG and metallic coated steel for packaging in Spain.

5.2.2. Ilva

5.2.2.1. The Ilva assets subject to the Transaction

(101)    With its Taranto plant, Ilva is the EEA's largest single-site integrated steel production facility (35). Ilva's history dates back to 1911 when six independent steel companies established the Ilva Consortium for the management of their plants, which covered the whole Italian national production of pig iron. Ilva S.p.A and other Ilva Group companies were acquired by the Riva family in 1996 from the Italian State, which previously controlled these companies through the Istituto per la Ricostruzione Industriale (‘IRI’).

(102)    Ilva’s main production site in Taranto, opened in 1964, is located in the South of  Italy next to the deep sea port of Taranto. Ilva has two other production plants in  the North of Italy, in Genoa and Novi Ligure, as well as six steel service centres (three in the plants, plus Marghera, Paderno and Legnaro). These other facilities do not have internal HR production capability but process HR provided by the plant in Taranto.  In France, Ilva Group also has a steel service centre (Tillet), and a welded tube production site (Socova) (36). Figure 10 shows the location of Ilva's facilities. (37) Those facilities are all part of the Transaction and constitute part of the Ilva assets.

(103)    The Ilva assets have a technical blast furnace capacity of […] and slabs capacity of almost […]. However, due to the restrictions to which it is currently subject, this capacity is temporarily limited to […]. The HR coil production capacity of the Ilva assets amounts to […] but it is currently effectively limited at […] due  to  the capacity restriction upstream. At its current level of capacity, the Taranto plant accounts for circa […]% of the total EEA production capacity for HR. (38) The Taranto plant is capable of producing not only crude steel but also a number of downstream products, including QP, HR, CR and GS. The steelworks also has the capability to produce welded tubes.

(104)    The plant in Taranto has favourable production conditions due to its economies of scale, its close location to a seaport and relatively modern facilities. This is illustrated in Figure 11 (39), which places the EEA production plants in relation to the age and size of their blast furnace.

Figure 11 – Age and size of the blast furnaces of flat carbon steel plants in EEA

[…]

(105)    The Ilva assets can produce and supply a wide range of flat carbon steel products. Figure 12 shows the wide product range of the Ilva assets in the flat steel market with the percentage of production share and the industries using those products.

Figure 12 – Ilva's products and production share (40)

[…]

5.2.2.2. Events leading to the sale of the Ilva assets

(106)    In the past years, Ilva has faced events leading to a reduction in its production volumes. In particular,  the  HR  production  area  in  the  Genoa  plant  was  closed  in 2005. In 2008, the Puglia region, where the Taranto plant is located, enacted a Regional Law against dioxins, which imposed limits for industrial emissions starting from 2009. In 2009, Ilva closed certain production facilities. (41)

(107)    The blast furnaces (42) of Taranto have been running at around […]% of their nominal capacity (three blast furnaces working out of five) since 2013 due to environmental constraints imposed on Ilva. In July 2012 the public prosecutor of Taranto subjected Ilva's local HR production lines to temporary seizure due to environmental issues. The Italian Government issued Law Decree no. 207/2012, (43) laying down conditions to issue an Integrated Environmental Authorisation (Autorizzazione Integrata Ambientale – ‘A.I.A’), to restart production activity. Under such authorisation, a production cap has been imposed to reduce emissions.

(108)    The former management of Ilva Group, namely the Riva family, had to step aside  and  was  replaced   by   a   government-appointed   Extraordinary   Commissioner   in June 2013. (44) The Extraordinary Commissioner had a mandate to ensure the continued industrial activity of Ilva. The Extraordinary Commissioner of 2013 designed an environmental plan to prevent further pollution and upgrade the Taranto plant in compliance with the environmental permit (the ‘2014 Environmental Plan’) (45) , which was approved on 14 March 2014 by the Council of Ministers. According to the 2014 Environmental Plan, the investment needed to bring the Taranto plant in line with the requirements of the environmental permit amounts to EUR 1.75 billion.

(109)    The 2014 Environmental Plan further constrained the production levels of  Ilva, which has started suffering losses at EBITDA level in 2012 after many positive years (excluding 2009 when certain production plants were closed).

(110)    Table 2 shows the total shipments of Ilva S.p.A. (reduced due to the production caps imposed and plant closures in 2009) for the years 2004–2017, including ArcelorMittal's plan for 2018, and selected financial figures of Ilva S.p.A in those years.

Table 2 – Ilva S.p.A's shipments and selected financial figures 2004 -2017 (46)

[…]

(111)    As described in recital (107), Ilva S.p.A has operated at around […]% of its nominal capacity since 2013, […] (47). For instance, according to a competitor of Ilva:‘Steel production involves significant sunk and fixed costs. Not producing is usually more expensive than producing. That is because the steel assets require significant capex [capital expenditure] to build and their maintenance, even if standing idle, is expensive. There are thus significant economies of scale that push to producing at high utilisation levels, particularly at the liquid and hot stages of the production.’ (48)

(112)    At the same time, the lower production level could not support the necessary environmental investments and the fixed costs. Besides operational inefficiencies, it appears that the main drivers of the operational loss have been the unproportioned fixed costs and the loss in spread on steel sales. The fixed costs have remained  mostly unchanged. Labour costs represented […]% of fixed cost in 2014 and 2015,  as calculated based on available data for 2014 and 2015. (49)

(113)    The operating result of Ilva S.p.A. (50) further suffered due to the decreased spread on steel sales, which decreased from around EUR […] from 2004 onwards to around EUR […] in 2013. Figure 13 from Ilva Group's business plan in May 2014 shows the historical development of the HR coil spread (the difference between average HR  coil selling price and raw material basket cost) for the period of 2000-2013, and the expected development of such spread from 2013 onwards. The decreased spread coupled with the production cut significantly affected Ilva S.p.A's performance, as it incurred substantial losses.

Figure 13 – Historical hot rolled coil spread 2000-2013 (51)

[…]

(114)    In June 2014, the Italian Government explored a potential interest of ArcelorMittal in acquiring Ilva Group. In November 2014, […], ArcelorMittal submitted a non- binding offer for certain assets of Ilva Group […], (52) but the assets were not  eventually sold by the Italian government. In January 2015, Ilva Group was admitted to extraordinary administration.

(115)    Under Law Decree no. 1/2015, an ad hoc insolvency procedure for Ilva Group was introduced. (53) Subsequently, Law Decree no. 191/2015 entitled ‘Disposizioni urgenti per la cessione a terzi dei complessi aziendali del Gruppo Ilva’ provided for the sale of Ilva Group's assets through a transparent and non-discriminatory public procedure.

(116)    On 5 January 2016, the Extraordinary Commissioners of Ilva Group published a call for expressions of interest in relation to the transfer of businesses owned by Ilva Group. Twenty-five interested parties were admitted to the preliminary due diligence phase, following which acquisition consortia were formed for the submission of formal offers by 30 June 2016. On 30 June 2016, two bidders were admitted to the final phase of the tender process: (i) the AM Consortium (as described in recital (9)) and (ii) the AcciaItalia Consortium (‘AcciaItalia') formed, inter alia, by the Italian steelmaker Arvedi, the Indian steel company Jindal Steel and the State-owned investment bank Cassa Depositi e Prestiti (‘CDP’).

(117)    On 26 May 2017, Ilva Group's Commissioners recommended to award the Ilva assets to the AM Consortium, (54) which offered a price of EUR 1.8 billion. (55) The Italian Minister of Economic Development issued the final adjudication decree in favour of the AM Consortium on 5 June 2017. (56)

5.3. Sources of supply

(118)    In  the  EEA,  flat  carbon  steel  is  supplied  by (i)  EEA-based  integrated suppliers,

(ii) EEA-based non-integrated suppliers and (iii) non-EEA suppliers (imports).

(119)    Integrated suppliers are active throughout all or most of the flat carbon steel value chain, and they are capable of producing semi-finished products (slabs) and HR internally. Integrated suppliers are thus generally not dependent on external sourcing of HR for their downstream production. Integrated EEA-based suppliers of flat carbon steel products include, for instance, ArcelorMittal, the Ilva Group, Arvedi, ThyssenKrupp, Tata Steel Europe, Salzgitter, Voestalpine and SSAB.

(120)    Non-integrated suppliers (sometimes also referred to as ‘re-rollers’) are only active in the production and supply of one or more flat carbon steel products downstream of HR. These suppliers thus depend on third parties for their supply of HR, which they process to downstream products for instance by re-rolling the HR into CR. Some of the non-integrated suppliers may be integrated between the downstream products, for instance in the production of all of CR, HDG and OC. Non-integrated EEA-based producers include, for instance, Marcegaglia and Wuppermann.

(121)    Non-EEA supplies consist of imports of flat carbon steel produced outside the EEA. They are considered further in Section 5.3.3.

5.3.1.     EEA-based integrated suppliers other than the Parties

(122)    The locations of the main integrated flat carbon steel producers are shown in Figure 14.

Figure 14 – Locations of main EEA integrated suppliers

[…]

(123)    Within the EEA, the majority of integrated producers are located in Central and Northern Europe, including ThyssenKrupp and Salzgitter primarily in Germany, Tata Steel in the Netherlands and in the UK, Voestalpine in Austria, US Steel Kosice in Slovakia, and SSAB in Finland and Sweden. In contrast, only Ilva Group, ArcelorMittal and Arvedi have major production assets in Southern Europe.

5.3.1.1.       ThyssenKrupp

(124)    ThyssenKrupp Steel Europe (‘ThyssenKrupp’) is headquartered in Germany. ThyssenKrupp is one of Europe’s major flat carbon steel producers and it is active throughout the flat carbon steel value chain from primary steel production to coated finished products.

(125)    ThyssenKrupp operates a number of production sites in the EEA, mostly in Germany where also all of its primary steelmaking capacity is located. According to the Notifying Party, ThyssenKrupp produces for instance: (57)

·             HR in four plants in Germany (Duisburg-Bruckhausen, Duisburg-Beeckewerth, Bochum and Hagen) with a total production capacity of ca. […] and with a product portfolio that […].

·             CR in five plants in Germany (Duisburg-Bruckhausen, Duisburg-Beeckewerth, Bochum, Dortmund and Andernach), including both commodity and high- value CR.

·             GS in eight plants. Seven of these plants are located in Germany (Duisburg- Bruckhausen, Duisburg-Beeckewerth, Bochum, Dortmund, Finnentrop, Kreutzal Eichen, and Kreutzal Ferndorf), and one in Spain (Sagunto). ThyssenKrupp supplies a wide range of products, […].

(126)    The locations of ThyssenKrupp’s production plants, as submitted by the Notifying Party, are shown in Figure 15.

5.3.1.2.       Tata Steel

(127)    Tata Steel (‘Tata’) is headquartered in India and has operations in 26 countries worldwide. Tata is one of the major EEA flat carbon steel producers and it is active throughout the flat carbon steel value chain from primary steelmaking to coated finished products.

(128)    In the EEA, Tata operates a number of production sites in the UK, the Netherlands, Belgium and Northern France. Its primary steelmaking sites with blast furnaces are located in the UK and in the Netherlands. According to the Notifying Party, Tata produces: (59)

·             HR in four production sites, three located in the UK (Brinsworth, Port Talbot, and Llanwern) and one located in the Netherlands (Ijmuiden), with total capacity of circa […].

·             CR in six plants, four located in the UK (Brinsworth, Llanwern, Port Talbot, and Trostre), one in the Netherlands (Ijmuiden) and one in France (Louvroil). Tata’s CR production range includes […].

·             GS in eight plants, four located in the UK (Llanwern, Cookley, Swansea and Shotton), two in the Netherlands (Ijmuiden and Moerdijk), one in Belgium (Liège) and one in France (Louvroil). Tata’s range focuses on […].

(129)    The locations of Tata’s production assets, as submitted by the Notifying Party, are shown in Figure 16.

5.3.1.3. Arvedi

(130)    The Italy-based and family owned Arvedi Group, founded in 1963, is active throughout the flat carbon steel value chain from primary steelmaking to coated finished products. Arvedi employs the EAF route for its steel production.

(131)    According to the Notifying Party, Arvedi operates one production site in Cremona,

Italy with a HR production capacity of ca. […]. The site also produces CR and GS. (61)

(132)    The Arvedi Group was part of the AcciaItalia Consortium. According to Arvedi, the Consortium planned to increase the  production  output  of  the  Ilva  assets  and  reach 10 Mt of steel production annually. Moreover, the AcciaItalia Consortium planned to close down two of the Ilva assets' ovens and instead use the DRI technology, which use gas instead of coal, as it expected that gas prices would be lower in the future. (62)

5.3.1.4.Other integrated EEA suppliers

(133)    In addition to the suppliers referred to in Sections 5.3.1.1 to 5.3.1.3, there are a number of other integrated and non-integrated suppliers of flat carbon steel with production in the EEA. These include:

(134)    Salzgitter AG (‘Salzgitter’) is an integrated steel producer headquartered in  Salzgitter, Germany. Salzgitter operates several steel plants in the EEA, three of which are dedicated to flat carbon steel production and are located in the neighbourhood of Salzgitter. According to the Notifying Party, Salzgitter has HR production capacity of approximately […] and it also produces CR and GS. (63)  The locations of Salzgitter’s production sites, as indicated by the Notifying Party, are shown in Figure 17.

(135)    VoestAlpine, is an integrated steel producer based in Austria. VoestAlpine operates five steel plants in the EEA, four of which are dedicated to flat carbon steel production, all located in Austria. According to the Notifying Party, VoestAlpine has a HR production capacity of circa […] and it also produces CR and GS, all with product portfolios […]. The locations of VoestAlpine’s production sites, as indicated by the Notifying Party, are shown in Figure 18.

(136)    SSAB-Ruukki is an integrated steel producer headquartered in Sweden and with EEA steel  production  plants  in  Sweden  and  Finland.  According to the Notifying Party, SSAB-Ruukki has a HR production capacity circa […] with a product portfolio […]. SSAB-Ruukki also produces CR with a […]. The locations of SSAB-Ruukki’s EEA production sites for finished flat carbon steel, as indicated by the Notifying Party, are shown in Figure 19. (66)

(137)    NLMK Europe operates, according to the Notifying Party, six steel production sites  in the EEA, located in France (Strasbourg and Beautor), Belgium (La Louvière and Clabecq), Denmark (Frederiksvaerk) and Italy (Verona) as shown in Figure 20. According to the Notifying Party, NLMK Europe’s HR production plant is located in Belgium (La Louvière), with capacity of circa […] with a […]. NLMK Europe also produces commodity CR and GS in the EEA. The locations of NLMK's EEA production sites for finished flat carbon steel, as indicated by the Notifying Party, are shown in Figure 20. (68)

(138)    Further EEA suppliers of flat carbon steel include, for instance: ISD, a Ukraine- based steel producer that operates production sites in Poland (Częstochowa – QP capacity) and Hungary (in Dunaújváros, where it operates under the name Dunaferr – HR, CR and GS capacity), with total HR capacity of circa […] (70); US Steel, a US- headquartered steel producer with operations in North America and Central Europe.  It operates one flat steel production plant in the EEA, located in Košice, Slovakia with HR, CR, GS, TP and OC capacity (‘USSK’); and Wuppermann, a Dutch- headquartered steel manufacturer with production facilities in Austria (Judenburg  and Linz), Hungary (Győr-Gönyű) and the Netherlands (Moerdijk). Wuppermann focuses on GS products. (71)

5.3.2.   EEA-based non-integrated suppliers

5.3.2.1. Marcegaglia

(139)    Marcegaglia is Ilva’s largest customer, and one of the leading steelmakers in the South of Europe. It has an annual output of about 5.4 million tonnes of steel products. (72) However, different from Ilva Group and ArcelorMittal, Marcegaglia is a non-integrated re-roller and does not have its own HR (or slabs) production capacity.

(140)    Marcegaglia, established in 1959, is Europe’s largest non-integrated steel supplier processing mainly carbon steel but also stainless steel in its rolling mills and SSCs. Processing steel accounts for 90% of the Marcegaglia Group’s turnover (73), the remaining 10% being generated from six other business areas: building, home products, engineering, energy, tourism and services.

(141)    The Marcegaglia Group is headquartered in Gazoldo degli Ippoliti, Mantova in Northern Italy and employs 5 400 people directly (mostly in the EEA). Marcegaglia has 16 steel processing plants (including SSCs) in Italy (74), and 6 other steel plants around the world (UK, Poland, Russia, USA, Brazil and China). Marcegaglia has a minority stake in a steel mill in Bremen, Germany, which is owned by ArcelorMittal (‘AM Bremen’).

(142)    Marcegaglia covers the steel value chain downstream from HR with its product offerings as Figure 21 shows.

Figure 21 – […] (75)

[…]

(143)    Marcegaglia's plants dedicated to the production of flat carbon steel products are located in Italy in Albignasego (CR lines), Gazoldo – Mantova (CR lines),   Ravenna (CR lines, GS lines, OC lines) and S.Giorgio Nogaro - Udine (QP lines) (76). Marcegaglia’s processing plant at Ravenna, on Italy’s northern Adriatic coast, is one of the largest steel processing site in the EEA capable of processing over […] of coil products a year. The main import port of Marcegaglia is also located in Ravenna, from where it distributes steel for further processing to other Marcegaglia plants within Italy. The port can handle ships of 40 thousand DWT (77).

5.3.2.2. Business model of Marcegaglia

(144)    Apart from Marcegaglia, all the most significant rerolling facilities in the EEA are owned by the integrated steel manufacturers. While there are some other independent re-rollers active in the EEA, particularly in Italy and Germany, those companies are typically niche players focussed on a specific grade and market segment that have significantly smaller capacities than Marcegaglia (78). In this respect, and  due to  its size, the business model of Marcegaglia is unique in the EEA.

(145)    Due to its size, and its lack of primary steel production capability, Marcegaglia is the largest independent HR coil consumer in the EEA, and one of the largest in the  world.

(146)    Historically Marcegaglia was a SSC operator and a tube manufacturer. Marcegaglia’s entry into the production of CR, GS  and  OC  products  only  took  place  around year 2000. Thus, Marcegaglia has already moved upwards in the integration ladder even if it has not acquired HR production capacity (79).

(147)    Marcegaglia observed that in order to produce higher quality products, stability in the supply chain is necessary. In order to achieve this, and to ensure a steady supply of a certain volume of HR, Marcegaglia has decided to focus on long term structural industrial agreements instead of investing in a steel mill (80).

(148)    Marcegaglia has agreements to ensure the supply of steel for processing  which arrives largely in the form of HR, but also as slabs (81). Besides Ilva, ArcelorMittal is the other important EEA supplier of Marcegaglia, with annual supply of around […], mostly supplied from ArcelorMittal's Bremen plant, in which Marcegaglia also holds a minority shareholding. Apart from Ilva Group and ArcelorMittal, Marcegaglia sources some HR coil volumes from other EEA suppliers, as well as from imports. Marcegaglia's sourcing from EEA suppliers other than ArcelorMittal and Ilva Group is more limited, due to those other suppliers' more limited capacity (‘the lack of available HR volumes’), focus on other geographic regions (‘sells them[HR] in northern Europe and not in Italy’), or the large proportion of captive use (‘ThyssenKrupp, Voest Alpine and SSAB offer no real volumes limiting their feeding to the captive market’ (82).

(149)    Marcegaglia is a significant importer of HR to Italy and to the EEA. It sources in the range of 40-60% of its need from EEA suppliers, while the rest is sourced from imports.

(150)    Even though Marcegaglia imports substantial volumes of HR, it holds the view that ‘it is not easy to import significant quantities from other countries worldwide.  Firstly, many markets are getting more regional with local producers giving priority to their domestic/regional markets. …As a matter of fact, Marcegaglia in 2016 and even more in 2017 needs to buy slabs and roll them into HRC through a hire-rolling agreement, in order to support its needs; such agreements are only temporary and volatile. They cannot represent a solid source in the long term. For many other  users, smaller than Marcegaglia, it is not easy to source [import] from everywhere because of size, service requirements and financial reasons.’ (83)

5.3.2.3.  […]

(151)    Marcegaglia acquired its minority shareholding of […] in AM Bremen […]:

Table 3 – […] (84)

[…]

(152)     […] (85).

(153) With regard to re-rolling of slabs, Marcegaglia explained (86) that the ‘indispensable factors to make it become a feasible option are: good established networks with slab providers; stability of the mill and flow required (e.g. Marcegaglia has in place very regular contract with Ilva for theoretically converting slab to HR up to […] of HR); possibility to finance a long cycle that results in higher costs; and logistic optimization (e.g. running a vessel with a capacity of at least […] tonnes).’

(154)     […]. (87)

(155)     […] (88).

5.3.2.4.  Marcegaglia's participation in the AM Consortium

(156)     Marcegaglia Carbon Steel S.p.A. is participating in the AM Consortium […]. (157)        […].

(158)     […].

5.3.3.  Imports and anti-dumping measures

5.3.3.1. Imports of flat steel products into the EEA

(159)    As mentioned above in Section 5.3, the production and supply of flat carbon steel products in the EEA is covered by integrated steel producers based in the EEA, non- integrated suppliers based in the EEA and non-EEA based suppliers (imports). This Section provides a brief overview of (i) the role imports play as a source of supply for EEA customers of flat carbon steel products; (ii) the flows of imported flat  carbon steel products; and (iii) historical developments regarding imports of flat carbon steel products into the EEA.

(160)    First, the Commission notes that the EEA flat carbon steel industry is to some extent affected by steel trade flows with non-EEA countries, that is imports and exports of flat carbon steel products. EEA-based steel producers export certain volumes of their production to countries outside the EEA. At the same time, non-EEA steel manufacturers from different countries export certain flat carbon steel products into the EEA. As such, EEA customers of flat carbon steel products may source these products not only from EEA-based steel producers, but also from producers based outside the EEA. The sourcing from imports differs and is to some extent limited by certain non-price related factors. Those factors will be specifically discussed in Sections 9.4.5.8 and 9.7.5.4.

(161)    Second, the flows of imported flat carbon steel products vary significantly year to year and are affected by numerous macro-economic and micro-economic factors. Historical data shows that imports of flat carbon steel products flow into the EEA     in 'waves', that is periods with high levels of imports  are  followed  by  periods  when the flow of imports is reduced and vice versa. Such waves are illustrated in Figure 22, which shows the imports of all flat carbon steel products for the time period 2005-2016.

(162)    Moreover, Figure 23 and Figure 24 depict such import flow 'waves' specifically with respect to HR, respectively into the EU and into Italy. It can be observed that for example the period between 2006–2008 was characterised by high levels of import flows, followed by a period of significant reduction in imported HR, and then again followed by an increase in imports in 2010–2011. Then, the period 2012–2014 is again a period of relatively low import flows followed by an increase in imports in the period 2015–2016 (for 2017 complete annual data are not available).

(163)    Third, imports are also affected by anti-dumping measures as described in the next Section. In recent years, there have been several anti-dumping measures (provisional or definitive) that have been imposed by the Commission on specific non-EEA producers. The purpose of these trade defence measures is to remove  injurious effects  of  dumped  or  subsidised  imports  on  the  Union  industry  producers   that produce the like products. Such trade measures might further affect the attractiveness of imports as EEA customers might become liable to pay higher duties for their imported material, or imports may be diverted away from the EEA, when trade defence measures are imposed.

5.3.3.2.       The European trade defence regime

(164)    At present, certain flat carbon steel products entering the EU are subject to anti- dumping duties. The effect of the anti-dumping duties on the competitive environment in the specific product markets  is  assessed  in  more  detail  in  Sections 9.4.5.7, 9.5.5.3 and 9.7.5 of this Decision. However, this Section aims to provide a brief overview of the EU's anti-dumping and anti-subsidy regime, while the next Section describes the trade defence measures currently imposed on specific products falling within the product markets concerned by this Decision.

(165)    The Commission is responsible for investigating allegations of dumping by exporting producers from non-EU countries. The general procedural rules applicable to anti- dumping investigations as well as the substantive rules for imposing anti-dumping duties are set out in Regulation (EU) 2016/1036 of the European Parliament and of the Council.- (91) Article 1(1) of that Regulation states a general principle whereby 'an anti-dumping duty may be imposed on any dumped product whose release for free circulation in the Union causes injury.'

(166)    The Commission may initiate an investigation after receiving a written complaint submitted by any natural or legal person, or any association not having legal personality, acting on behalf of the Union industry. Such complaint needs to be supported by a sufficient representation of the Union producers and sufficient evidence that dumped or subsidised imports originating in non-EU countries are causing or threatening to cause material injury to a Union industry. In special circumstances, the Commission may initiate an investigation on its own initiative. (92)

(167)    Once an anti-dumping investigation is launched, the Commission has maximum      15 months to conduct its investigation and finalise its findings in anti-dumping cases and 13 months in anti-subsidy cases. (93)

(168)    The scope of the Commission's anti-dumping investigation consists in determining whether (i) the product under investigation ('the product concerned') originating in the countries concerned is being dumped or subsidised; (ii) the dumped imports have caused injury to the Union industry; and (iii) it is in the wider Union interest to impose measures.

(169)    Moreover, the finding  of  dumping  and  subsidisation  covers  a  defined  period  (the 'investigation period') while the examination of trends relevant for  the assessment of injury, causality and Union interest covers longer time periods, typically four years. In the context of the Union interest test, the Commission examines whether the interest of the Union industry in being relieved from the injury caused by the unfairly traded imports is outweighed by the burden the trade defence measures would create for the downstream market, including importers and users  (for example customers). Typically, the Commission carefully examines the impact of the planned measures on the cost structure of the importer and the user industry on the basis of verified questionnaire responses from the downstream companies concerned.

(170)    If the Commission finds that the required conditions for imposing trade defence measures are met, the Commission may impose measures on the product concerned. The measures may take a number of different forms, including an ad valorem duty, specific duty or minimum import price. The Commission may also accept price undertakings from specific companies.

(171)    Even before an adoption of definitive anti-dumping measures, the Commission may decide to adopt provisional anti-dumping measures. Article 7 of Regulation (EU) 2016/1036 empowers the Commission to impose provisional duties if certain conditions are met, including the condition of a provisional affirmative determination having been made of dumping and consequent injury to the Union industry and of Union interest. Any such provisional duties must be imposed no earlier than 60 days and no later than nine months from the initiation of the proceedings. (94)

(172)    Trade defence measures are imposed for a period of five years and may, in some circumstances, be subject to review. However, unless an expiry review is initiated, a definitive trade defence measure will expire after five years. (95)

5.3.3.3. Trade defence measures currently in place and ongoing anti-dumping investigations by the Commission

(173)    The Commission has conducted a number of anti-dumping investigations with regard to different steel products in recent years. As a result of these investigations, there  are currently definitive trade defence measures in place on HR products originating  in China as well as Brazil, Iran, Russia and Ukraine. Moreover, definitive measures have also been imposed on CR from China and Russia. Furthermore, the  Commission also recently adopted definitive anti-dumping measures for corrosion resistant steel products, which fall within the HDG product type, coming into the EU from China.

(174)    First, with regard to HR products, on 13 February 2016 the Commission initiated an anti-dumping proceeding concerning imports of certain hot-rolled flat products of iron, non-alloy or other alloy steel originating in China. The Commission first imposed provisional anti-dumping measures on these products on 6 October 2016, effective from 8 October 2016. (96) The provisional measures are listed in Table 4.

(175)    Subsequently, on 5 April 2017, the Commission adopted definitive measures on the concerned HR products from China, effective from 7 April 2017. (97) The definitive measures are listed in Table 5.

(176)    The definitive duties were subsequently amended by Commission Implementing Regulation (EU) 2017/969. (98) The amended duties were supplemented by countervailing duties. Both are listed in Table 6 and Table 7.

(177)    In  addition  to  the  anti-dumping  investigation  relating  to  HR  from  China,  on    7 July 2016 the Commission initiated an anti-dumping proceeding concerning  imports  of  certain  hot-rolled  flat  products  of  iron,  non-alloy or  other  alloy steel originating in Brazil, Iran, Russia, Serbia and Ukraine. No provisional measures were imposed in this case. On 5 January 2017 imports of the concerned HR products from Russia and Brazil have been made subject to registration. (99) As a result of the Commission's investigation, on 5 October 2017 definitive anti-dumping measures were adopted with regard to HR products originating in Brazil, Iran, Russia and Ukraine, effective from 7 October 2017, whilst the investigation on imports of HR from Serbia was terminated. (100) The definitive measures on HR from Brazil, Iran, Russia and Ukraine are listed in Table 8 and Table 9.

(178)    Second, with regard to CR products, on 14 May 2015 the Commission initiated an anti-dumping proceeding concerning imports of certain cold-rolled flat steel products originating in China and Russia. These products were subsequently made subject to registration on 11 December 2015. (101) On 10 February 2016 the Commission imposed provisional anti-dumping duties on the CR products concerned. (102) The provisional measures, effective from 13 February 2016, are listed in Table 10.

(179)    Following the provisional measures and further investigation, definitive measures on CR from China and Russia were imposed on 29 July 2016. (103) The definitive duties, effective from 5 August 2016, are listed in Table 11 and Table 12.

(180)    In addition to the definitive duties in that investigation, the Commission also concluded that the conditions for retroactive application of the definitive anti- dumping duty were met. Therefore, a definitive anti-dumping duty was levied on the product concerned that was made subject to registration by  Implementing  Regulation (EU) 2015/2325. (104) Indeed, in addition to imports which caused injury during the investigation period, further significantly increased volumes of the  product concerned had been imported after initiation, at prices even lower  than during the investigation period. These large and low priced quantities of the product concerned have already had a further bearing on the prices and Union market share  of the Union industry.

(181)    Third, with regard to HDG products, on 9 December 2016 the Commission initiated an anti-dumping proceeding concerning imports of certain corrosion resistant steel products originating in China. On 7 July 2017, these corrosion resistant steel  products have been made subject to registration. (105)  Subsequently, on 9 August  2017, the Commission imposed provisional anti-dumping duties on these products. (106) The provisional duties, effective from 11 August 2017, are listed in Table 13.

(182)    Following the provisional measures and further investigation, definitive measures on corrosion   resistant   steel   products   originating   in   China   were   imposed   on    7 February 2018. (107) The definitive measures, effective from 9 February 2018, are listed in Table 14.

(183)    As the trade defence measures on certain HR, CR and corrosion resistant steel products cover some of the major steel producing and steel exporting countries, any assessment of the extent to which imports of HR, CR and HDG products may exert competitive pressure on EEA-based flat carbon steel producers and, in particular the merged entity post-Transaction, must be made in light of the situation as restored by anti-dumping duties. The Commission's assessment of the imports' role in the HR,  CR and HDG product markets is presented in Sections 9.4.5, 9.5.5 and 9.7.5 respectively.

5.4. Pricing mechanism in the flat carbon steel industry

(184)    As transactions in flat carbon steel markets relate to differentiated products sold through different channels under different contractual/commercial arrangements, prices and pricing mechanisms are equally differentiated. However, given that, notwithstanding all these differences, the output essentially flows from the same primary steelmaking capacities and hot strip mills, prices for different flat carbon steel products may be significantly interdependent.

(185)    In the first place, spot prices for commodity flat carbon steel serve as a general industry benchmark, as they reflect the market equilibrium between the prevailing level of demand, on the one hand, and the supply conditions, including available capacity and raw material prices, on the other hand. While commodity flat carbon steel sold on the spot markets accounts for only a portion of the overall output, it also drives the prices of non-commodity products in spot sales or longer term supply arrangements.

(186)    In spot transactions, most operators use the so-called base price plus extras pricing model. (108) In this model, the price is formed by combining a base price for the basic steel product (commodity grade, dimension, finishing etc) and the fees charged for ‘extras’, that is features adding more value to the basic product, such as specific grade quality (for example alloys), dimensions (for example thin gauges), surface quality, finishing (coating type such as the ‘zinc extra’ for GS) and services. Whereas producers tend to change their base prices regularly in response to market evolutions, their price lists for ‘extras’ are modified less frequently (for example on a yearly, or an even longer basis).

(187)    Therefore, a change in a supplier's base price will not only impact commodity products, but also specialty/non-commodity products, the price of which is determined as an extra over the base / commodity prices. Therefore, commodity products exert a competitive constraint on specialty grades, in such a way that if prices of commodity grades decrease significantly (because the base price decreases), prices of specialty grades will also tend to decrease as a result.

(188)    This is largely confirmed by flat carbon steel producers who participated in the market investigation, which for instance stated that: (109)

(a)   ‘Price for high-end products is intertwined with that of commodity grades. The price of high-end grades is computed on the basis of a reference price, which corresponds to the price of the most basic commodity grade, by adding specific price extras required for instance in terms of required dimension (thickness  and width), type of packaging, etc. For each flat steel product, like hot rolled, cold rolled and galvanized, the basic steel grade used as a reference is different. Usually the producers publish the price extra list […] Typically the negotiation with customers relate to the reference price only. This price system is used not only with steel service centres, but also with all types of customers.’

(b)   ‘Changes in commodity prices also affect the price of high quality products’.

(189)    The indices reporting prevailing spot prices are published periodically (daily,  weekly, monthly) by market intelligence agencies, such as MEPS and SBB/Platts, which regularly update the price indices based on a representative sample of spot transactions reported by the operators. The price index aggregates the base prices of suppliers, which are established by deducing from the spot price the costs of extras. The price indices may be used in negotiating spot transactions as well as supply agreements, and may even form a component of the price setting mechanism in such agreements. (110)

(190)    In addition, the demand for commodity products may influence the production and supply of non-commodity products, and vice versa. Along with the arbitration between merchant sales and captive use, (111) such mutual influence may rebalance capacity allocation for different flat carbon steel products produced in the same plant, and consequently impact prices. For example, an oversupply of commodity steels, and the consequent drop in price, may, in order to maintain a cost efficient  production scale in their plants, lead producers to increase production of non- commodity/specialty products and thus drive the prices down for these products. Conversely, a stronger demand for specialty steels, for example GS for automotive applications, may cause the supplier to reallocate part of its production from commodity HR to higher grade HR used in the production of automotive GS, thus meeting stricter product specifications. However, such reallocations of production between commodity and non-commodity products is subject to certain limitations as explained in Section 6.10.

5.5. Demand trends

(191)    As countries and economies develop, their steel consumption per capita increases. The higher the importance of manufacturing, investment in fixed assets and urbanization in a given economy, the higher the consumption of steel. Once the peak point of per capita consumption is reached, the consumption starts declining and stagnates. Once stagnation is reached, the demand for steel appears to be cyclical and linked to the GDP evolution of economies. These trends can be well observed in Figure 25 which shows the evolution of steel demand worldwide for 1950–2017 (112). For example, from 1995 onwards in China there has been a boom in steel demand (industrialisation/modernisation of China), while the growth in steel demand in Europe, US and Japan was very moderate.

(192)    Around 50% of the steel produced worldwide goes into steel buildings and infrastructure (rail, bridges, tunnels, housing, transport) and 13% is consumed by the automotive sector. This consumption is dependent on the performance of the economy and the business cycles of these industries. Consequently, demand for steel is cyclical and shocks in the world economy, like for instance the 2008 financial crisis, strongly impact steel demand as, for example big infrastructure projects are  cut back or delayed, and the consumers tend not to buy new cars and invest in new homes.

(193)    Concerning the demand trend for the next 8 to 10 years, as shown in Figure 26, steel demand in Europe is expected to grow at approximately […]% per year till 2025, from […] in 2015 to […] in 2025 (113). The same trend is true worldwide, where steel demand is also expected to increase on average […]% per year, following the dip in steel demand in 2015 (mostly due to lower demand in China).

Figure 26 – Expected evolution of global steel demand 2015-2025

[…]

(194)    The same trend is supported by the analysis of UBS Global I/O, according to which ‘the EU steel demand remains robust, driven by a still strong auto[motive] and continuous recovery in the construction sector.’ (114)

6.  PRODUCT MARKET DEFINITION

6.1. Production and supply of semi-finished flat carbon steel products

(195)    In its past decisions, the Commission found that semi-finished carbon steel products are distinct from finished carbon steel products, and considered potential segmentation of semi-finished steel products based on basic shapes into: (i)   blooms; (ii) billets; and (iii) slabs. (115) Slabs are typically used to produce flat products whereas blooms and billets are primarily employed in the manufacturing of long products.  The present case only concerns flat carbon steel products.

(196)    The Notifying Party does not contest the segmentation explored in past Commission decisions.

(197)    The Commission considers that the precise scope of the relevant product market may be left open for the purposes of assessing the Transaction, as the Transaction would not significantly impede effective competition in the production and supply of semi- finished flat carbon steel products regardless of the product market definition.

6.2. Production and supply of finished flat carbon steel products

6.2.1.  Overview of the markets

(198)    In past decisions, the Commission has distinguished steel products based on the chemical composition of the steel (metallurgical characteristics) on the one hand, and according to the physical shape of the products on the other. Based on chemical composition, the Commission has distinguished four broad categories of steel products: (i) carbon steel, (ii) stainless steel, (iii) specialty steel and (iv) electrical steel. (116)  The present case only concerns carbon steel.

(199)    As regards the physical shape of carbon steel products, the Commission has in the past held that flat products constitute distinct product markets, as opposed to long carbon steel products. (117)

(200)    Finished flat carbon steel products are produced from semi-finished products, in particular slabs. There are three main stages in the production process of finished flat carbon steel products: (i) hot-rolling, (ii) cold-rolling and (iii) coating. The finished products may be sold at the end of each of those processes and, in line with those production stages, the Commission has considered in previous decisions that the following finished  flat  carbon  steel  products  constitute separate  product  markets:

(i)   quarto plates, which are produced in specific quarto (four-stand) mills (‘QP’),

(ii)   hot-rolled products excluding quarto plates (‘HR’), (iii) cold-rolled steel   (‘CR’),

(iv) galvanised steel (‘GS’), (v) metallic coated steel for packaging; and (vi) organic coated steel (‘OC’). (118)

6.2.2.The Parties’ activities

(201)    The Parties’ activities overlap in the production and supply of (i) semi-finished flat carbon steel products (slabs), and finished flat carbon steel products, namely: (ii) QP, (iii) HR, (iv) CR (v) GS, including both hot-dip galvanised (‘HDG’) and electro- galvanised steel (‘EG’) and (vi) metallic coated steel for packaging, including in both tin plate (‘TP’) and electrolytically coated chromium steel (‘ECCS’).

(202)    ArcelorMittal also produces OC while Ilva does not.

(203)    In addition to the production and supply of primary flat carbon steel products, the Parties are active in a number of downstream activities, such as steel distribution and tube-manufacturing.

6.3. Production and supply of hot rolled flat carbon steel

(204)    Hot rolling is a process whereby hot slabs are rolled into coils (HR) in a hot strip mill or into flat quarto plates (QP) in a quarto mill. The latter are non-coiled products  with very different dimensions, in particular in terms of thickness, from other hot- rolled flat carbon steel products.

(205)    The Commission has in previous cases considered that HR and QP constitute distinct relevant product markets. (119)

(206)    The Notifying Party agrees with the Commission’s previous practice, which the market investigation has not called into question.

(207)    Therefore, considering all information available to it and for the purposes of the present case, the Commission finds that (i) HR products, excluding QP, and (ii) QP constitute distinct relevant product markets.

6.4. Production and supply of cold rolled flat carbon steel

(208)    CR products can be delivered ‘full-hard’ or annealed. Annealing refers to a heat treatment to the CR coil that improves its machinability. According to the Notifying Party, supplies of full-hard CR in the merchant market are limited and they constitute […]% of the total CR sales.

(209)    The Commission has previously defined a single relevant market for all uncoated CR carbon steel flat products. (120)

(210)    ArcelorMittal agrees with the Commission's previous practice. (121) It further submits that full-hard and annealed CR are simply different stages in the production process but do not constitute different products and should be considered as belonging to the same product market.

(211)    The market investigation has confirmed the negligible role of full-hard CR in the market and has in general not called into question the Commission’s previous practice. (122)

(212)    Therefore, considering all information available to it and for the purposes of the present case, the Commission finds that all CR products form part of the same relevant product market.

6.5. Production and supply of galvanised flat carbon steel

6.5.1. Introduction

(213)    As explained in Section 5.1, GS may be obtained by two main production  processes:

(i) hot-dip galvanisation and (ii) electro-galvanisation. The Parties’ activities overlap in the production and supply of both hot-dip galvanised flat carbon steel products (‘HDG’) and electro-galvanised flat carbon steel products (‘EG’). The Transaction gives rise to affected markets regardless of whether HDG and EG are considered separately or together.

(214)    HDG products are produced through uncoiling and reheating the steel strip and feeding it through a bath of molten metal composed of zinc or zinc and  other elements at an appropriate temperature. The dipping process results in a metallic (zinc-based) coating on the steel substrate.

(215)    EG products are produced through the application of an electrolytic coating process which results in a zinc-containing coating on one or both sides of the strip.

(216)    In addition to the traditional production methods, the Notifying Party produces GS using a proprietary technology known as jet vapour depositation (‘JVD’). According to the Notifying Party, JVD enables the application of zinc coatings on any steel grade in a low pressure chamber, […]. Ilva does not produce any products using  JVD.

6.5.2. Previous cases

(217)    In a number of early cases dating to the 1990s, the Commission suggested that HDG and EG may constitute distinct markets, in view of their different product characteristics and prices. The Commission nonetheless left the exact product market definition ultimately open in those cases. (123)  Further, in two cases dating to 2006    and 2007, the Commission acknowledged the two different types of GS but only had to carry out an assessment on the basis of HDG. (124)

(218)    However, in Usinor/Arbed/Aceralia, which is from 2001, the Commission concluded that the relevant product market encompasses all GS products, namely both HDG  and EG. The Commission substantiated its conclusion, for instance, by referring to developments in the quality of HDG that enabled it to be used in applications where previously only EG had been used. The Commission further observed that, based on the information it had at its disposal in that case, the price differences between HDG and EG were not substantial, noting that the official price lists of suppliers indicated the average difference in price between EG and HDG sheets to be below 5%. (125)

(219)    In Mittal/Arcelor, which is from 2006, and in ArcelorMittal/Galvex, from 2008, the Commission again concluded that the relevant product market comprises all GS and found no distinct markets for HDG and EG, noting that the market investigation had confirmed such approach. (126) Later, the Commission has followed the approach in a number of cases but has not assessed the relevant market definition in detail. (127)

6.5.3.The Notifying Party’s views

(220)    In this case, the Notifying Party submits that all GS products belong to the same relevant product market, irrespective of whether they are HDG or EG. This also applies to GS produced through JVD which, according to the Notifying Party  belongs to the overall GS market. However, in case HDG and EG were to be considered to constitute distinct markets, the Notifying Party considers that JVD is closer to EG than HDG because both EG and JVD have thinner zinc coatings compared to HDG.

(221)    According to the Notifying Party, there is both supply-side and demand-side substitutability between HDG and EG.

(222)    With regard to supply-side substitutability, the Notifying Party submits that all major EEA steel producers produce both EG and HDG.

(223)    With regard to demand-side substitutability, the Notifying Party submits that technological development in the quality of HDG has progressively eliminated the most relevant differences between HDG and EG. According to the Notifying Party, HDG and EG are today interchangeable from the customers’ perspective in all key aspects and they are substitutable for a wide range of different applications.

(224)    The Notifying Party further submits that HDG is replacing EG in the market: HDG products are generally cheaper than EG products and hence customers are steadily switching to HDG from EG and, simultaneously, suppliers are shutting down EG production lines. […]. The Notifying Party’s submission on the development of production capacities is shown in Figure 27.

Figure 27 – ArcelorMittal's estimate of GS capacity development (128)

[…]

(225)    According to the Notifying Party, the fact that it has so far continued to produce EG despite […] compared to HDG does not put the substitutability between HDG and  EG into question. Similarly, […]. (129)

(226)    In addition, the Notifying Party suggests that the prices and margin spreads for EG and HDG move together, further supporting the finding of an overall market for GS with no sub-segmentation into EG and HDG.

6.5.4.  The Commission’s assessment

6.5.4.1.  Framework for assessment

(227)    The Commission recalls that market definition is a tool to identify and define the boundaries of competition between firms. The objective of defining a market is to identify  the  actual  competitors  of  the  undertakings  involved  that  are  capable of constraining those undertakings’ behaviour and of preventing them from behaving independently of effective competitive pressure. (130)

(228)    To define the relevant market, the Commission will typically consider both supply and demand-side substitutability and may also take into account other  evidence which shows competitive constraints (or lack thereof) between different products and their producers. (131)

6.5.4.2.  Supply-side substitutability is at most limited between HDG and EG

(229)    The Commission recalls that, when assessing supply-side substitutability, it is essential to consider whether firms are able to switch production between the products in the short term and without incurring significant additional costs or risks, in response to small but permanent changes in relative prices. (132) Therefore, while the fact that a number of suppliers only produce one of the goods under assessment may be an indication that the products do not belong to the same product market, the opposite is not necessarily true. In other words, it is not sufficient that a number of suppliers produce both products under assessment. It needs to be determined whether suppliers actually could (and would) switch production between the products in response to small but permanent changes in relative prices.

(230)    The Commission observes that HDG and EG are produced through different production processes, and they require different production lines and machinery. Unlike, for instance, many different grades of HDG that can be produced on a given HDG line, HDG and EG cannot be produced on the same production line.

(231)    The results of the market investigation purport that suppliers would not switch any volumes or would only switch limited volumes between HDG and EG in case of a small but significant price increase of one of them compared to the other. (133) A steel producer explains that ‘due to the different production processes and assets needed for producing either EG or HDG, it is not possible to switch production from one product to the other if there is no spare capacity’ while another one considers that they ‘need to utilize the existing production line for EG’ (134)

(232)    Furthermore, while a number of suppliers considered that EG is in the process of being replaced by HDG, most of the suppliers that took a position considered this to only apply to certain applications but not all. A supplier explains: ‘OEM/automotive sector is replacing EG by HDG. In some applications, however, mainly in the electrical engineering sector (computers) and in other industrial applications, EG will still be used’. Another supplier concurs and notes that EG is ‘still requested by customers’. (135)

6.5.4.3.      Demand-side substitutability is limited between HDG and EG

(233)    The results of the market investigation purport that demand side substitutability is limited and, in particular, EG is not readily substitutable with HDG. A clear majority of distributor customers taking a position indicated that they would either not switch or only switch less than 10% of their sourcing of EG to HDG even if EG saw a small but significant price increase compared to HDG. (136) While a number of distributor customers noted that their choice is ultimately dictated by their clients, one of them explained that ‘HDG and EG are two different qualities you cannot switch – EG will be painted afterwards while HDG is already protected by the zinc’. (137) As to  industrial customers, the majority of them agreed with the Notifying Party in that HDG is suitable for applications where particularly good surface quality is needed, for instance in the automobile industry. (138) Nonetheless, a clear majority of those taking a position, including also one automobile manufacturer, nonetheless indicated that they would not switch or would only switch less than 10% of their EG sourcing to HDG even if EG saw a small but significant price increase compared to HDG, including also one automobile manufacturer. (139) Some industrial customers also referred to technical reasons preventing or hampering switching from EG to HDG; they explained for instance that ‘HDG is complexe [sic] for welding’, ‘[o]ur technical department have some technical concerns if they had to switch from EG to HDG - - risk of tools dammage [sic]’ and ‘[o]n Ceilings products, only the EG surface is adequate to do post painting’. (140)

(234)    The results of the market investigation were largely similar also when assessing switching from HDG to EG. In particular, a clear majority of both distributors and industrial customers indicated that they would not switch or would only switch less than 10% of their sourcing of HDG to EG even if HDG saw a small but significant price increase compared to EG. (141)

6.5.4.4.      Price correlation analysis provided by the Notifying Party is inconclusive at best

(235)    The Notifying Party has submitted that the prices and spreads for HDG and EG move together. The Notifying Party’s analysis is shown in Figure 28 and Figure 29.

Figure 28 – Evolution of EG and HDG Prices (142)

[…]

Figure 29 – Evolution of EG and HDG spreads (143)

[…]

(236)    However, the Commission notes that, in general, price correlation analysis can only provide indirect evidence of market definition, and is more suitable as a ‘separation’ test rather than an ‘inclusion’ test. That is to say, it can provide insights on whether two products do not belong to the same relevant market, rather than whether they belong to the same relevant market. This is because the evidentiary value of price correlation analysis is highly dependent on controlling for common cost and demand factors. (144)

(237)    The Commission  further  notes  that  the  analysis  provided  by  the  Notifying  Party shows instances where prices and spreads take different patterns. For instance, Figure 28 and Figure 29 show a change in the series for Italy where the difference between the price and spread noticeably changes in 2016.

(238)    Furthermore, the results of the market investigation suggest that the prices for EG  and HDG are not always dependent on each other. To this effect, the Commission observes that while several customers, in particular distributors, indicated in the market investigation that the prices for EG and HDG follow each other, the majority of customers (including also distributors) indicated that they are not priced according to the same indices. Moreover, the Commission observes that a number of the respondents who considered that the prices for EG and HDG follow each other appear to refer to common cost components, such as the price of zinc or the price of the substrate steel input. (145) A reference to zinc and steel substrate was also made by the only steel supplier that considered in the market investigation that the prices of EG and HDG follow each other. (146)

(239)    While some of the industrial customers observed, for instance, that EG tends to follow the pricing of HDG, a clear majority of industrial customers and steel  suppliers taking a position indicated that the prices for EG and HDG do not follow each other and that the same indices are not followed in their pricing. (147) An industrial customer explains in this respect: ‘HDG is only available in the European market to a certain extent. There has been a dramatic price increase (approx. 30%) in the last 18 months with regard to HDG. The situation is different for EG: there is over- capacity’. (148)

6.5.4.5.      Price differences between HDG and EG are significant

(240)    Actual price differences between HDG and EG are significant. For the purpose of facilitating   a   ‘like-for-like’   comparison,   the   Notifying   Party   has    submitted information on the ex-works prices of EG and HDG in the EEA in 2016. The information is shown in Table 15.

(241)    The Commission observes that, based on the information provided by the Notifying Party, EG was on average […] more expensive than HDG in 2016. This is also more than the difference in production costs ([…]) as provided by the Notifying Party. (151) Moreover, the Commission notes that the Notifying Party has not provided information for the price of EG of less than 0.5 mm; if that thickness is excluded from the calculation, the price difference between HDG and EG appears to be even higher.

(242)    In Usinor/Arbed/Aceralia in 2001, the Commission noted that the average price differentials in the Community between EG and HDG sheets were below 5% according to the official price-lists of the main suppliers. (152) […].

6.5.4.6.  ArcelorMittal’s internal documents and behaviour suggest limited substitutability between HDG and EG

(243)    The Commission observes that, in its ordinary business documents, ArcelorMittal […]. (153) Contrary to the submission of the Notifying Party, the Commission considers that this supports the view that HDG and EG products are not fully substitutable.

(244)    The Commission further observes that ArcelorMittal has continued to produce EG under conditions that suggest it would not have done so were HDG and EG adequately substitutable. […]. (154) […]. (155)

(245)    Based on the information available, it does not appear logical that ArcelorMittal would have continued to produce all the volumes of EG it has produced if EG could be substituted with HDG from both the demand and supply side as submitted by the Notifying Party.

6.5.5.  Conclusion

(246)    On balance and considering all evidence available to it, the Commission considers that there is at least a serious possibility that the production and supply of EG and of HDG constitute distinct product markets. However, as the outcome of the  competitive assessment of the Transaction remains the same and the Commitments entered into by the Notifying Party address the Commission’s concerns under all alternative market definitions, it is not necessary for the Commission to conclude on the exact product market definition.

6.6. Production and supply of organic coated flat carbon steel

6.6.1.  Previous cases

(247)    In its past decisions, the Commission has considered that organic coated flat carbon steel products constitute a separate product market from other flat carbon steel products. (156)

6.6.2. The Notifying Party's views

(248)    The Notifying Party agrees with the approach taken by the Commission in its past decisions, whilst claiming that the precise scope of the relevant product market for OC in the current case may be left open, as the Transaction raises no competitive concerns regardless of the market definition.

6.6.3.The Commission's assessment

(249)    The Commission observes that OC products can be produced on the basis of both a GS substrate and CR substrate. However, the Commission notes that the facts of the Transaction, together with the evidence presented by the Notifying Party, do not give reason for the Commission to reconsider the approach taken in its past decisions as regards the product market definition of OC.

6.6.4. Conclusion on the product market for the production and supply of organic coated flat carbon steel products

(250)    The Commission finds that, for the purposes of the assessment of the Transaction,  the exact definition of the relevant product market for the production and supply of organic coated flat carbon steel products can be left open as the Transaction would not significantly impede effective competition in relation to OC regardless of the market definition.

6.7. Production and supply of metallic coated steel for packaging

(251)    The production of metallic coated steel for packaging involves coating a thin, flat carbon cold-rolled coil or sheet with a fine layer of metal – tin for tin plate (TP) or chromium for electrolytic chromium coated steel (ECCS). Before undergoing the tinning process, CR strips are usually annealed and tempered to improve ductility  and increase strength, respectively. Chromium coating consists of an electrolytic treatment in a bath which deposits metallic chromium and chromium oxide on the blackplate.

6.7.1. Previous cases

(252)    In its past cases, the Commission has defined a separate relevant product market for steel for packaging, leaving open a potential distinction between beverage and non- beverage steel packaging applications. (157)

6.7.2. The Notifying Party's views

(253)    First, the Notifying Party argues that the relevant product market may be defined as TP, without further subdivision.

(254)    Second, the Notifying Party submits that from a supply-side perspective, the production routes for the manufacture of steel packaging for beverage cans may also be used for the manufacturing of steel cans for non-beverage packaging. (158)

(255)    Third, the Notifying Party submits that TP and ECCS belong to the same relevant product market because of the significant supply-side and demand-side substitutability. (159)

(256)    As regards demand-side substitutability, the Notifying Party argues that TP can be used in all applications for which ECCS can be used. However, as acknowledged by the Notifying Party, due to the fact that ECCS cannot be easily wall ironed and welded, it cannot be used in certain applications for which TP is suitable, such as aerosol containers and welded or wall ironed cans. Nevertheless, the Notifying Party argues that these applications for which ECCS is not suitable represent a minority of total supply. (160)

(257)    From a supply-side perspective, ArcelorMittal argues that the manufacturing processes for TP and ECCS are almost entirely the same. The difference consists in the type of metallic bath used during the coating phase, containing tin for TP and chromium for ECCS. According to the Notifying Party, in order to switch production from one product to the other, a steel manufacturer only needs to employ different electrolytes, deposition cells, tanks, and piping in the electroplating process. Thus the same production line can produce both TP and ECCS. (161)

6.7.3. The Commission's assessment

(258)    During the market investigation of the Transaction, the Commission received feedback from market participants both with regard to supply-side and demand-side considerations for metallic coated steel for packaging. This feedback concerned potential limitations in terms of substitutability between TP and ECCS and their suitability for certain applications, namely beverage packaging and non-beverage food packaging.

(259)    From a supply-side perspective, a majority of respondents to the Commission's investigation active in the supply of flat carbon steel products in the EEA stated that they cannot produce steel for beverage packaging and steel for non-beverage packaging on the same production lines. (162) When asked about the production of TP and ECCS, a majority of respondents active in the production and supply of flat carbon steel in the EEA considered that they cannot produce TP and ECCS on the same production line. (163)

(260)    From a demand side perspective, the customers' feedback received during the Commission's market investigation indicated that there are certain differences between beverage and non-beverage packaging which may also affect the steel needed for the purpose. (164) Similarly, customers reported on certain limitations that may affect substitutability of TP with ECCS and vice-versa.

(261)    First, it has been reported that the thickness of the metal, mechanical properties (hardness and tin coating) and roughness (finish) are different in beverage and non- beverage packaging. (165) These differences may pose limitations as to the customers' ability to switch sourcing from steel for beverage packaging to steel for non-  beverage packaging and vice-versa. (166) A customer noted: 'In the food packaging sector, and in particular for the products that [the company] produces, the quality of tinplate has to be premium. Steel for non-food packaging can be of lower quality.' (167)

(262)    Second, packaging comes in different shapes, which affects the type of steel that can be used for its production. As a customer active in the metal packaging industry explained, there are certain differences in the manufacturing process of three-piece cans and two-piece cans. For instance, ECCS cannot be used in order to produce cans that require welding. (168)

(263)    Third, concerning the substitutability of ECCS by TP, the market investigation pointed out that technically such a switch would be possible. However, it is not practical. This is not only because TP is more expensive, but also because it would lead to a different visual appearance of the caps, which would be different due to the substrate of the metal beneath. (169)

(264)    Fourth, concerning the substitutability of TP by ECCS, a majority of respondents that source steel for packaging noted that they cannot switch sourcing from TP to ECCS, or only switch limited volumes. (170) Besides the differences in the product characteristics and customer requirements, it was pointed out that ECCS may not be acceptable from a quality perspective due to potential corrosion and the fact that it is not weldable, thus making it unsuitable for certain applications. (171)

6.7.4. Conclusion on the product market for the production and supply of metallic coated steel for packaging

(265)    The Commission finds that, for the purposes of the assessment of the Transaction,  the exact definition of the relevant product market for metallic coated steel for packaging can be left open as the Transaction would not significantly impede effective competition under any product market definition, even if TP and ECCS belonged to separate product markets.

(266)    Moreover, the Commission finds that for the purposes of the assessment of the Transaction, it is not necessary to conclude whether a distinction should be made based on the application, namely between the production and supply of metallic coated steel used for beverage packaging and the production and supply of metallic coated steel used for non-beverage packaging, could affect the product market definition. The Transaction would not significantly impede effective competition even if separate product markets are defined based on the application.

6.8. Production and supply of welded carbon steel tubes

6.8.1. Introduction: main delineations in previous cases

(267)    In its previous decisions, the Commission has considered various different delineations of the product markets for the production and supply of steel tubes on the basis of the type of steel used in the tubes, the size of the tubes and their technical characteristics.

(268)    The Commission has previously distinguished tubes made of carbon steel from tubes made of other types of steel. (172) Further, the Commission has distinguished small- diameter tubes from large-diameter tubes given different production processes  and fields of application. (173) In previous cases the Commission has also distinguished  welded tubes from seamless tubes, given different production processes,  properties, fields of application and prices. (174) Within the welded carbon steel tube market, the Commission has further distinguished between standard tubes and precision tubes. (175) Furthermore, within standard tubes, the Commission has considered various potential sub-segmentations such as structural tubes / hollow sections, commercial tubes and mechanical tubes. (176)

6.8.2.  Small welded carbon steel tubes

6.8.2.1.  Previous cases

(269)    The Commission has also considered the segmentation of small welded (including standard and precision) tubes markets into separate sub-markets for hot-formed and cold-formed tubes. (177) This is due to the fact that both standard carbon steel tubes and precision carbon steel tubes can be either cold-finished or hot-finished.

(270)    The present case concerns small welded carbon steel tubes, in particular small welded standard carbon steel tubes and small welded cold-finished tubes.

6.8.2.2. The Notifying Party's views

(271)    The Notifying Party generally agrees with the Commission's previous product segmentations. However, ArcelorMittal does not agree with any further sub- segmentation of the market for small welded carbon steel tubes into hot-formed small welded tubes and cold-formed small welded tubes, as this does not reflect the critical supply-side consideration that all manufacturers are capable of producing all types of tubes.

6.8.2.3. The Commission's assessment

(272)    The replies to the market investigation support the Commission's previous practice in distinguishing between tubes made from carbon steel and tubes made from other types of steel: small diameter tubes from large diameter tubes as well as welded tubes from seamless tubes. The replies also support the Commission's previous practice in distinguishing between standard/commodity tubes and precision tubes. (178)

(273)    Whilst in more recent cases, the Commission explored a further segmentation of the market between hot-formed and cold-formed small welded tubes, it ultimately left the question open. (179) Respondents to the preliminary market investigation in these proceedings indicated that supply and demand substitutability may be limited. (180)

(274)    The replies to the preliminary market investigation support the view that there is limited supply side substitutability between cold and hot formed small welded carbon  steel tubes even if a number of suppliers can produce both. A majority of competitors who responded to the investigation indicated that they do not consider that the know-how and equipment required for the manufacture of hot-formed small welded tubes is similar to that required for the production of cold-formed small welded tubes. Whilst similar production equipment is used for both, the additional cold forming process needs additional knowledge and production equipment. Furthermore, all competitors who responded consider that hot and cold-formed small welded tubes are not comparable in price. This is due to the different production processes involved, as there are different costs involved.

(275)    Demand side substitutability was also indicated to be limited. A majority of customers indicated that they could not switch or could only switch a small part of their demand from hot-formed to cold-formed small welded tubes. (181) The reasons given as to why it would be difficult to switch include customers' technical requirements (related to  safety), regulatory requirements and pricing issues. Responses were similar when customers were asked whether they could technically switch demand from cold-formed small welded tubes to hot-formed small welded tubes. A majority of respondents indicated that they could not switch or could only switch small parts of their volumes. Reasons given against switching include pricing (20% price difference), customers' requirements and technical differences.

(276)    Furthermore most competitors who responded to the investigation indicated that there are price differences between hot and cold-formed small welded tubes. (182)

6.8.2.4. Conclusion on the product market for the production and supply of small welded carbon steel tubes

(277)    The Commission finds however that the question as to the exact product market definition can be left open, as the Transaction would not significantly  impede effective competition regardless of whether small welded carbon  tubes  are considered as one market or are segmented into different distinct markets.

6.8.3.   Large welded carbon steel tubes

6.8.3.1.  Previous cases

(278)    In previous decisions, the Commission has distinguished between large diameter tubes and smaller diameter tubes, due to differences in their production processes and fields of application. (183) The Commission left open the question whether longitudinally welded and spirally welded tubes also form a distinct relevant product market. (184)

6.8.3.2.  The Notifying Party's views

(279)    The Parties submit that both longitudinally welded and spirally welded tubes compete in almost all tenders and belong to a single product market, although spirally welded tubes are most commonly used in applications where pressure and service conditions are less onerous.

6.8.3.3.  Conclusion on the product market for the production and supply of large welded carbon steel tubes

(280)    The Commission finds that the question as to the exact product market definition can be left open, as the Transaction would not significantly impede effective competition regardless of whether large welded carbon tubes are considered as one market or are segmented into different distinct markets.

6.9. Distribution of flat carbon steel

6.9.1. Previous cases

(281)    In its past decisions, the Commission has considered separate markets for the distribution of steel products on the basis of the different distribution channels between  Steel  Service  Centres  ('SSCs'),  Stockholding  Centres  ('SCs')  and   Oxy-cutting Centres. The Commission also considered further segmentation of the distribution market according to groups of products sold: (i) carbon steel products/stainless steel products and (ii) flat steel products/long steel products. (185)

6.9.2.  The Notifying Party's views

(282)    The Notifying Party partially agrees with the Commission’s approach in previous cases.

(283)    With respect to SSCs, the Notifying Party submits that metallic coated steel for packaging is distributed through dedicated SSCs as the material cannot usually be processed by regular SSCs. With respect to distribution through SCs, the Notifying Party submits that, a further segmentation of the market based on groups of products is not appropriate, as SCs offer almost all varieties of steel products (including long and flat products of carbon and stainless steel) and act as wholesalers.

6.9.3. The Commission's assessment

(284)    The results of the market investigation do not call the Commission’s previous approach into question.

(285)    Furthermore, for the purposes of this case, it is not necessary for the Commission to assess the market definition in detail as the outcome of the competitive assessment of the case is not altered by the market definition retained.

6.9.4.  Conclusion on the product market for distribution of flat carbon steel

(286)    The Commission finds that it is not necessary to conclude on the exact  product market definition with regard to steel distribution as the Transaction would not significantly impede effective competition regardless of how the markets are defined.

6.10. Commodity products and specialty products

6.10.1.  Introduction

(287)    In terms of product quality, based on the Commission’s investigation, a distinction may potentially be made between commodity and non-commodity flat carbon steel products, with the latter also referred to as 'specialty' products. Whereas commodity products are typically supplied to the general industry and distribution, non- commodity/specialty steel products are supplied to specific industries, such as the automotive or aerospace industries, where they are employed in demanding applications where high-end types of steel (for instance high-strength but low  weight) are needed.

6.10.2.  Previous cases

(288)    In numerous previous cases, the Commission has not considered the distinction between commodity and non-commodity products within flat carbon steel. (186)

(289)    However, in the recent case M.7155 – SSAB / Rautaruukki, the Commission found that it was likely that high-strength steels (‘HS’) and wear resistant steel (‘WR’) steel could belong to a market separate from commodity flat carbon steel, though it left the exact market definition ultimately open. (187)

6.10.3. The Notifying Party's views

(290)    ArcelorMittal acknowledges that there is differentiation within flat carbon steel and that, in particular, Ilva concentrates on commodity products whereas ArcelorMittal has a wider product portfolio.

(291)    ArcelorMittal further submits that the boundaries between commodity and specialty products are unclear, as there is no specific or industry-wide definition of which products can be qualified as commodity or specialty. This is also the case with  respect to grades, which constitute just one of the many considerations that determine whether a flat steel product is commodity or specialty alongside factors such as physical dimensions and coatings.

6.10.4. The Commission’s assessment

(292)    The results of the market investigation support ArcelorMittal’s submission in that while a distinction between commodity and high-end products is recognised, there is no industry-wide accepted definition of each of them. For example, one competitor explained that ‘There is no generally accepted method for the distinction between commodity and high-end steel grades’. (188) Another major EEA steel manufacturer noted that ‘…high end products are those with specific mechanical and chemical properties, used mostly in the automotive and niche industries. Commodity products are those used in normal construction applications’. (189)

(293)    The results of the market investigation further purport that most EEA flat carbon  steel producers have the ability to produce commodity and high-end products on the same production lines. This applies to both primary as well as downstream rolling mills. (190)

(294)    Further, it appears that even if certain steel producers suggested that they prioritise high-end products where possible (191) the production of commodity and high-end products on a given production line are linked to each other. In fact, as also  suggested by ArcelorMittal, it appears very difficult for  a  steel  producer  to  dedicate 100% of its production capacity of a given production line to specialty products. Periods of ‘relaxation’ are required at all stages, from casting, through rolling, to finishing. (192) For instance, not all slabs derived from the same caster sequence can be used to produce specialty products; worn cylinders cannot be used  to roll specialty products whereas they can still be used to produce commodity products; and specialty HDG might require a zinc bath having specific conditions. This is corroborated by the statements made by certain producers that indicate that they resort to commodity products instead of high-end products if demand is not high enough or there is spare capacity. (193)

(295)    Therefore, the Commission considers that there is likely a noticeable level of supply- side substitutability between high-end and commodity carbon steel flat  products. This finding is not called into question solely by the fact that there are a limited number of suppliers – such as Ilva – that predominantly operate in the commodity products only.

(296)    Furthermore, in light of the evidence available to the Commission, commodity and high-end products constitute a continuum of products that differ to a varying extent  in their quality and properties. This suggests that also customers may, at least to some degree, arbitrate between different steel qualities within the continuum. (194)

(297)    Finally, the Commission observes that the present case does not concern any particular niche products that would call the Commission to make a more detailed analysis of supply- and demand-side substitutability between them and other flat carbon steel products.

(298)    Overall, the Commission considers that the available evidence supports the view that the markets for flat carbon steel are differentiated between commodity and high-end products but that those two groups of products do not, for the purposes of the present case, constitute distinct product markets within the various flat carbon steel markets.

6.10.5.  Conclusion

(299)    On balance and considering all evidence available to it, the Commission considers that distinct product markets should not be defined for commodity and high-end products within the markets for the production and supply of flat carbon steel products. There is, however, some degree of differentiation between commodity and high-end products which the Commission has taken into account in its competitive assessment of the Transaction.

7. GEOGRAPHIC MARKET DEFINITION

7.1. Production and supply of semi-finished flat carbon steel products

7.1.1.  Previous cases

(300)    In its past cases, the Commission has assessed semi-finished steel markets on an EEA-wide basis, (195) or both EEA-wide and worldwide basis, (196) but ultimately left the precise scope of the geographic market definition open.

7.1.2. The Notifying Party's views

(301)    The Notifying Party submits that the relevant market for semi-finished flat carbon steel products is at least EEA-wide, in particular in light of the absence of barriers to EEA trade and given price convergence throughout the EEA. At the same time, the Notifying Party submits that the precise scope of the relevant geographic market may be left open as the Transaction raises no competitive concerns regardless of the market definition. (197)

7.1.3.The Commission's assessment

(302)    The Commission observes that the outcome of the competitive assessment is not altered however the market is defined.

7.1.4. Conclusion on the geographic market for the production and supply of semi-finished flat carbon steel products

(303)    The Commission finds that for the purposes of the assessment of the Transaction, the exact definition of the relevant geographic market for the production and supply semi-finished flat carbon steel products can be left open as the Transaction would not raise competitive concerns on this market, however defined.

7.2. Production and supply of finished flat carbon steel products

(304)    The Parties' activities overlap in the production and supply of finished flat carbon steel productions, namely HR, CR, GS (HDG and EG) and metallic steel for packaging (TP and ECCS).

(305)    Nonetheless, only the production and supply of (i) HR, (ii) CR and (iii) GS,  including HDG and EG, give rise to competition concerns (See Sections 9.4.7, 9.5.8, 9.7.7, 9.8.3 and 10.10). Therefore, the Commission's assessment of the relevant geographic market definition presented below in Sections 7.2.1 to 7.2.7 is focused on HR, CR and GS, including HDG and EG.

7.2.1. Previous cases

(306)    The Commission has in some previous cases considered that the geographic markets for finished flat carbon steel products could be Community / EEA-wide (if  not wider), though it has not always concluded on the exact geographic scope of these markets. (198) It should though be borne in mind that, at the time of the earlier cases, the area covered by the Community was smaller than what the EEA covers today.

(307)    In the most recent decision in which geographic markets were assessed in detail, M.7155 – SSAB/Rautaruukki, the Commission looked at the possibility that markets for certain finished flat carbon steels could be cross-border regional in scope within the EEA. In particular, the Commission concluded that there was at least a serious possibility that three Nordic countries (Finland, Sweden and Norway) constituted a market separate from the rest of the EEA at least for commodity HR, CR and OC.

(308)    In the present case, Ilva has all of its production facilities located in Italy while ArcelorMittal has production activities in numerous locations throughout the EEA. Geographically closest to Ilva are ArcelorMittal’s production facilities in Italy and Southern France. The Commission has recently assessed a number of transactions involving ArcelorMittal and flat carbon steel particularly in Southern Europe. However, in none of those cases was the Commission called to assess or conclude on the geographic market definition as no competition concerns arose under any alternative market definition.

7.2.2. The Notifying Party’s views

(309)    The Notifying Party submitted that the relevant geographic markets for flat carbon steels are at least EEA-wide, and that the markets should not be delineated according to regions within the EEA. Further, the Notifying Party maintains that focusing on sub-EEA markets (such as Southern Europe) would run counter to the global nature of the steel industry.

(310)    The Notifying Party supported its submission by explaining, for instance, that there are no barriers to intra-EEA trade, that prices are correlated throughout the EEA, transport costs are limited and that major suppliers and customers have an EEA-wide footprint. ArcelorMittal further explains that its own sales and marketing strategy is EEA-wide.

(311)    The Notifying Party also maintained that Northern and Southern European prices and spread are highly correlated suggesting that price differences between the two areas cannot persist as they are counteracted by flows from one region to another. In this respect, ArcelorMittal also submits that flows from northern suppliers to the South of Europe are significant and together with imports from outside the EEA account for the majority of sales in that region.

(312)    The Notifying Party submitted that flat steel demand is not spread homogeneously in Southern Europe as it is rather concentrated in the North of Italy and the North of Spain. This would imply that suppliers located in Northern Europe would be situated as close or closer to these areas as Ilva or as other southern flat carbon steel producers. Following this reasoning, ArcelorMittal further submits that northern suppliers constrain southern suppliers and that a 5–10% price increase brought about by southern producers would not be profitable as northern suppliers would increase sales to such an extent as to make the price increase unprofitable.

(313)    This alleged reaction by northern suppliers would be driven by two main elements: the level of transport costs and the spare capacity of northern producers. To that effect, ArcelorMittal submits that transport costs are not of material relevance for inter-regional flows throughout the EEA: First, regardless of any price increase in Southern Europe, most plants located in Northern and Central Europe face similar transport costs to transport volumes to Southern Europe as do plants located in Southern Europe. Second, in case of a 5–10% price increase the transport cost advantage of southern producers would be cancelled out. Furthermore, according to the Notifying Party, northern suppliers have ample spare capacity and would be able to increase their sales to Southern Europe following a hypothetical price increase there.

(314)    Furthermore, the Notifying Party submitted that a number of other EEA suppliers are equally well positioned to supply Southern Europe as the Parties are, including in particular the areas where most of the merchant market demand in the Southern European region is concentrated, that is Northern Italy and Northern Spain (which account for […]% and […]% of the Southern European demand for HR,  respectively, according to the Notifying Party). (199)

(315)    The Notifying Party observes that a number of suppliers located in Germany and even in the Benelux are within a similar distance from Northern Italy and Northern Spain as Ilva’s Taranto steelworks and many of ArcelorMittal’s plants are. These include, among others, ThyssenKrupp and Salzgitter from Germany, Tata from the Netherlands and Voestalpine from Austria, which is geographically even closer to Northern Italy than Ilva’s Taranto steelworks is. According to the Notifying Party, these suppliers could and would profitably supply customers in Southern Europe if prices there were to increase.

(316)    Furthermore, the Notifying Party submitted that transport costs are not prohibitive and would not prevent suppliers located outside the Southern European region from supplying there. In particular, the Notifying Party considers that a hypothetical price increase of […] would cancel any transport cost advantage that the Parties may currently enjoy when supplying to Northern Italy, which is the main steel consumption location in Southern Europe. The Notifying Party’s assessment of the transport costs and of a hypothetical price increase is shown in Figure 30.

Figure 30 – […] (200)

[…]

7.2.3.             Framework for assessment

(317)    The Commission recalls that a relevant geographic market refers to a geographic area in which the merging companies offer their products and 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 areas. (201)

(318)    In its assessment, the Commission will take into account various factors, including past evidence of diversion of orders to other areas, basic demand characteristics, views of customers and competitors, current pattern of purchases, trade flows as well as barriers to trade and costs associated to diverting orders to suppliers located in other areas. (202)

(319)    For the reasons set out in more detail in Section 7.2.4, the Commission finds that it is not appropriate to consider markets wider than the EEA. Moreover, as explained in Section 7.2.5, several elements indicate that the structure of supply and demand for flat carbon steel products and the conditions of competition are differentiated  between different parts of the EEA.

7.2.4. The markets are not wider than the EEA

(320)    ArcelorMittal has submitted that the markets for flat carbon steels are ‘at least EEA wide’ in scope, and that pricing for flat carbon steel products is driven by global dynamics.

(321)    The market investigation has not revealed evidence that would allow  the Commission to consider the markets to be wider than the EEA. In particular, conditions of competition do not appear homogeneous within and outside the EEA, market structures largely differ across different geographic regions and sourcing occurs to a very large extent at a regional level. A number of market participants have also indicated that imports cannot fully substitute EEA supplies. (203)

(322)    Further, the Commission considers the pricing of flat carbon steel products to be significantly differentiated across different geographic areas around the world as prices are not only affected by global development in raw material and global demand/supply balances but are also significantly affected by domestic factors that drive local price differentiation. Different developments in price spreads over a common index for raw material costs are observed in Figure 31. As that figure shows price spreads in the US and Japan move significantly differently as compared to the European series (North and South) and to Turkey and China. European prices might be more closely associated with prices in Turkey and China, however significant differences can be observed also across these regions.

Figure 31 – Price spreads over raw material costs for selected countries (HR, 2009-2017) (204)

[…]

(323)    The differentiated pricing across different geographic areas in the world, including between the EEA and neighbouring countries, such as Turkey can also be observed from market reports as well as ArcelorMittal's internal documents.

(324)    For example, an SBB daily briefing published on 22 January 2018 mentions, among others, that domestic and export HR coils (‘HRC’) prices began to show different tendencies in recent days. 'Turkish mills HRC export offer prices almost exceeded their domestic quotes, as demand is currently more lively in Europe.' (205) Furthermore, for instance [internal document]. (206)

(325)    Therefore, the Commission finds that it is not appropriate to consider the relevant geographic markets for flat steel products to be larger than the EEA.

(326)    Nonetheless, this does not entail that imports do not exert any competitive constraint on carbon steel markets within the EEA. Competitive constraints exerted by imports, where applicable and to the extent relevant, are taken into account in the competitive assessment of each relevant market.

7.2.5. Conditions of competition are at least differentiated within the EEA

7.2.5.1. Preliminary remarks

(327)    For the reasons indicated in the above Section 7.2.4, the Commission concludes that the EEA geographic market is sufficiently homogenous and can be distinguished from its neighbouring areas because the conditions of competitions are appreciably different in those areas. Nevertheless, the Commission considers that, for the reasons outlined in the following recitals, within the EEA there are variations in the competitive  conditions  that  need  to  be  taken  into  account  in  the competitive assessment. The Commission's findings on geographic variation are consistent with the Horizontal Merger Guidelines principles (207) and with the Commission practice. (208)

(328)    In particular for HR products, the Commission observes that the distribution of ArcelorMittal's activities is spread evenly across the EEA because of its large network of plants. Conversely, the other EEA suppliers' presence is more concentrated to the geographic area closer to their plants, as they generally own one primary steel production site. (209) Indeed, Ilva's sales are concentrated in Southern European countries, and in particular in Italy, as can be seen in Table 16, with more than […]% of HR sales made in Italy. For CR and GS Ilva also has meaningful sales to non-Southern countries, in particular to […] for HDG, although those sales  account for a limited part of the overall sales in those countries.

(329)    On this basis, the Commission notes that the Parties' activities overlap particularly in a number of Southern European countries, especially in Italy. Therefore, for the purposes of the present case, it is particularly instructive to consider whether or not the conditions of competition in the region where Ilva mostly sells and has noticeable market presence are differentiated from those present in other parts of the EEA.

7.2.5.2. Differentiation between Southern and Northern Europe suggested by market participants and demand trends

(330)    The results of the market investigation purport that competitors and customers alike recognise the differentiation in the conditions of competition between Southern Europe and Northern Europe.

(331)    With regard to customers, the majority of those taking a position indicate that the conditions of competition are different between Southern Europe and Northern Europe at least in some respects. While there is some variation in the replies depending on the product in question, the overall view is the same for HR, CR, HDG and EG. (212)

(332)    A market participant indicates: 'Currently, pricing levels in the North of Europe, where ArcelorMittal is active, are slightly higher compared to the South of Europe. Competition trends also appear to differ between these two regions since the distributors in the South of Europe tend to be more independent and  less constrained, especially due to the geographic location of Italy and Spain which favour inflows of imports. Specifically, the acquisition of Ilva by ArcelorMittal will remove one important steel operator which – together with other producers – contributes to preserve a certain degree of competitiveness in the South of Europe. The current level of competition in Southern Europe allows customers to source supplies for lower prices in this region compared to Northern Europe.' (213)

(333)    Another customer also indicates: '[…] can only notice that prices are typically lower in southern Europe, partly because of the existence of independent low-cost suppliers like Ilva. In Europe ArcelorMittal is currently a price leader.' (214)

(334)    In addition, a customer indicates: 'In addition, If Ilva ceases to act as a price constraint in southern Europe, there might be a convergence of prices between northern and southern Europe'' (215)

(335)    Further, another customer reports: 'In a post-merger scenario, customers in Spain  and Portugal will be affected due to the decrease in the number of potential suppliers. […] will probably play less of a role due to the effect of anti-dumping measures, therefore two of the main alternative players (ArcelorMittal and Ilva) would not compete, and the remaining competition would only be from imports'. (216)

(336)    The replies of flat carbon steel producers are more balanced. While not all competitors agreed, the majority of competitors taking a position confirm the replies from customers, indicating that the conditions of competition are different between Southern Europe and Northern Europe with regard to HR, CR, HDG and EG. (217) A competitor explains: ‘There is generally a higher level of competition in the Southern European market due to presence of a larger number of smaller players (including re-rollers). Integrated producers in Northern Europe, e.g. ArcelorMittal, ThyssenKrupp, Tata, Voestalpine, Salzgitter, have higher market share and generally stronger market power than the more fragmented producers in Southern Europe, especially in Italy’ (218)  while another supplier notes that price differences are caused by ‘[i]mport pressure over the Mediterranean Sea, higher transport cost from southern Europe to northern Europe, lower factor [sic] cost (wages, energy] in southern Europe’. (219)

(337)    The differentiation in the conditions of competition between Southern Europe and Northern Europe also finds support in internal documents of ArcelorMittal. […] (220) […]. (221)

(338)    In the market investigation, the Commission also sought to establish the boundaries of what market participants understand as the Southern European region when considering the conditions of competition. To that effect, a clear majority of customers and competitors taking a position indicated that Italy, Spain and Portugal belong to the Southern European region – in contrast to, for instance, Germany and the Benelux which most respondents considered not to belong to that region. Similarly, a majority of respondents taking a position also included Greece within Southern Europe but many respondents indicated that they did not have enough information to take a position. Similarly, also for Southern France the majority of customers taking a position considered it to belong to Southern Europe but many indicated they did not have enough information to take a position. On the contrary, most respondents taking a position considered Northern France not to belong to the Southern European region. The Commission observes that while there was some differentiation in the responses provided for different flat carbon steel products, this does not affect the overall position presented in this recital. (222)

(339)    With regard to the Southern European region suggested by market participants, the Commission observes that the majority of demand for steel in that area is found in Italy and Spain: As can be seen in Table 17, Italy and Spain represent between 80% and 90% of demand for each of HR, CR, HDG and EG when considering a Southern European region consisting of Italy, Spain, Portugal, Greece and Southern France.

(340)    Given the dominance of Italy and Spain in the demand for steel in the Southern European region, the majority of steel supplies in that region are directed to those countries. This pattern can also be observed in Ilva’s deliveries to the region as seen in Table 16.

(341)    The replies of market participants reported in this Section suggest that the competitive conditions are largely similar throughout the Southern European region.

7.2.5.3. The structure of demand is differentiated within the EEA

(342)    While different types of customers can be found throughout the EEA, it seems that the structure of demand and the customer base differs to some extent across different regions. With regard to Southern Europe, the customer base appears to be more fragmented than that in Northern Europe with a number of small distributors being present. (224) In addition, the re-rollers – first and foremost Marcegaglia – appear to  play a more important role in Southern Europe than in Northern Europe.

(343)    The difference in customer bases also seems to correspond to different types of products typically consumed by customers located in different regions of Europe: ArcelorMittal itself submits that customers in Southern Europe have a tendency of purchasing commodity products. That view has been confirmed in the market investigation where a number of flat carbon steel producers observed that consumption in Southern Europe is more focused on commodity products while demand in Northern Europe tends to be driven by different factors and consist more, in relative terms, of specialty products. (225)  That is also consistent with the finding that independent distributors and re-rollers, as opposed to sophisticated industrial end- users of steel, are more representative of typical customers in Southern Europe than in Northern Europe.

7.2.5.4. The structure of supply is differentiated within the EEA

(344)    The majority of integrated flat carbon steel suppliers are located in Central, Eastern and Northern Europe. Those include Thyssenkrupp and Salzgitter primarily in Germany, Tata Steel in the Netherlands and in the UK, Voestalpine in Austria, US Steel Kosice in Slovakia and SSAB in Finland and Sweden – as well as  ArcelorMittal in numerous locations including for instance Germany, Belgium, Poland and Northern France. In contrast, only Ilva, Arvedi and ArcelorMittal have significant production assets, including crude steel and hot-rolled flat carbon steel (HR) production in Southern Europe (Italy, Spain and Southern France). Ilva and Arvedi do not have manufacturing facilities outside Southern Europe

(345)    The locations of HR production sites, and the related capacities, as provided by ArcelorMittal, are shown in Figure 32.

Figure 32 - EEA HR production sites and HR rolling capacity (226)

[…]

(346)    The differences in the structure of supply do not only concern the location of manufacturers, but also the share of supply covered by imports. In particular, there are differences in import levels across EEA countries.

(347)    Based on information provided by ArcelorMittal, HR imports into the EEA flow mostly to Southern Europe where they account for a significantly higher share of supply than in Northern Europe and contribute also to the production of other downstream products such as CR and  HDG/EG.  The  flows  of  imports  for  CR and HDG are still sizeable in Southern European countries, but are more spread across the entire EEA. Belgium, for example, is one major importer of CR and HDG. Table 18 shows the imported volumes, sales share and sales distribution of HR, CR and HDG in a selected number of EEA countries.

(348)    The results of the market investigation support the finding that, in particular for HR, imports exert more competitive pressure in Southern Europe, including in Italy and Spain, than in Northern Europe. (229) A flat carbon steel producer, while noting that imports affect price levels throughout the EU, explains: ‘As the imports most consist of commodity products, their role in the more quality-driven North is less than in the South’  while  another  supplier  simply  submits  that  ‘[c]ompetitive  pressure  from non-EEA imports in southern Europe is higher’. (230) Similarly, a Spanish customer notes that ‘North market more controlled by EU Steel mills and its distribution. Less iimpacted [sic] by imports’. (231) Those comments are in line with further replies by customers, and in particular by a clear majority of SSCs customers, which have submitted that imports exert a stronger competitive constraint on commodity products. (232)

7.2.5.5.       Producers in the EEA tend to supply regionally

(349)    The Commission observes that producers generally supply flat carbon steel products close to their manufacturing facilities, and thus that the geographic location of the steel producers' plants appears to  significantly affect  their  supply patterns  across the EEA. Trade flows of flat carbon steels between Member States exist, and are sometimes sizeable. In particular for HR, however, it appears that producers sell the majority of their products in the area close to their respective production sites (which can encompass part of one or more Member States). This local differentiation is also reflected in the variation in market shares computed at different geographic level (Section 9.3.2).

(350)    First, Ilva makes the clear majority of its sales to customers in Italy where its production facilities are located. In 2016 Ilva supplied approximately […]% of its total HR sales to customers in Italy ([…]), while it only supplied around […]% of its total HR sales respectively to customers in […]. Ilva’s supply pattern for HR is shown in Figure 33.

Figure 33 – Ilva's HR sales distribution in the EEA, 2016 (233)

[…]

(351)    A similar distribution pattern is generally seen also with regard to Ilva’s supplies of CR and GS (HDG and EG). However, compared to HR, the patterns for CR and HDG are not as strongly Italy-centred. In particular, it appears that Ilva is selling a significant share of its HDG supplies also outside of Italy, in particular in […]. Ilva’s sales distribution of CR and HDG are shown in Figure 34 and Figure 35.

Figure 34 – Ilva's CR sales distribution in the EEA, 2016 (234)

[…]

Figure 35 – Ilva's HDG sales distribution in the EEA, 2016 (235)

[…]

(352)    Second, based on information supplied by ArcelorMittal, a localised distribution pattern is not only applicable to Ilva but can also be observed in the actual sales of a number of major EEA flat carbon steel producers. Those include, for example Salzgitter, Tata Steel and ThyssenKrupp, each of whom sold a significant amount of their steel production in the country where their production plants are located. Specifically, in 2016 Salzgitter sold […]% of its HR products in Germany, Tata sold more than […]% of its HR products in UK and the Netherlands, and ThyssenKrupp sold […]% of its HR products in Germany. The HR sales distribution of Salzgitter, Tata Steel and ThyssenKrupp is shown in Figure 36, Figure 37 and Figure 38 respectively.

Figure 36 – Salzgitter's HR sales distribution in the EEA, 2016 (236)

[…]

Figure 37 – Tata Steel's HR sales distribution in the EEA, 2016 (237)

[…]

Figure 38 – ThyssenKrupp's HR sales distribution in the EEA, 2016 (238)

[…]

(353)    In addition to HR, the regionalised sales patterns also apply, although to a lower extent, to the more valuable downstream products, such as CR and HDG. This concerns not only Ilva but also other major suppliers. For instance, in 2016, Salzgitter, which has its main production sites in Germany, supplied […]% of its CR and […]% of its HDG in Germany, while supplying minimal volumes of CR and HDG in other EEA countries. ThyssenKrupp, which also has its main production  sites in Germany (and a facility for the production of HDG in Spain), supplied […]% of its CR and […]% of its HDG in Germany, while supplying significantly lower volumes in other EEA countries. Tata Steel on the contrary seems to have more widespread sales of HDG across different countries. Figure 39 shows Salzgitter’s CR sales distribution and Figure 40 shows ThyssenKrupp's HDG sales distribution in   the EEA in 2016. Figure 41 shows Tata Steel's HDG sales distribution in the EEA in 2016.

Figure 39 – Salzgitter's CR sales distribution in the EEA, 2016 (239)

[…]

Figure 40 – ThyssenKrupp's HDG sales distribution in the EEA, 2016 (240)

[…]

Figure 41 – Tata Steel's HDG sales distribution in the EEA, 2016 (241)

[…]

(354)    Third, the Commission further observes that, while the volumes of steel consumption in each Member State vary and can affect the supply patterns, it does not alone explain the observed distribution patterns by EEA suppliers, because they are not proportional to the demand in each country. For instance, according to information supplied by the Notifying Party, the total HR deliveries in Germany by all suppliers were […] in 2016 (merchant market sales, including imports) while they were […] in Italy. Therefore, while the total merchant market demand in Italy was circa […]% higher than in Germany, Ilva’s supplies to Italy were over […] times its supplies to Germany – while, for instance, both ThyssenKrupp and Salzgitter supplied many times more HR to customers in Germany despite the total demand of HR being  higher in Italy.

(355)    Fourth, ArcelorMittal appears to be an exception, having relatively uniform distribution patterns across the EEA, as shown in Figure 42. That is likely due to its production facilities being spread across the EEA, as shown in Figure 32, allowing ArcelorMittal to supply its customers from a wide network of production locations and to be equally present across several Member States.

Figure 42 – ArcelorMittal's HR sales distribution in the EEA, 2016 (242)

[…]

(356)    Nevertheless, also for ArcelorMittal sales from a specific plant are concentrated in few regions. For example, Figure 42 shows ArcelorMittal's HR sales distribution from its plant in Aviles, Spain, that serves  in  large  part  customers  located  in  Spain ([…]%). Similarly, Figure 43 shows ArcelorMittal's HR sales distribution from its plant in Fos Sur Mer,  France, that serves in large  part  customers  located in   Italy ([…]%), Spain ([…]%) and France ([…]%). The regional sales patterns of ArcelorMittal's plants is also reflected by an internal document of ArcelorMittal about the plant of Fos-sur-mer, that only produces HR products, which mentions that […]. (243) It should be noted that such sales pattern likely depends on ArcelorMittal's subdivision of sales within its network. In this respect, it is logical for plants within a network to serve customers that are close-by rather than shipping to the area closer to another plant within the same network.

Figure 43 – ArcelorMittal's HR sales distribution from Aviles plant in the EEA, 2016 (244)

[…]

Figure 44 – ArcelorMittal's HR sales distribution from Fos Sur Mer plant in the EEA, 2016 (245)

[…]

(357)    Fifth, during the market investigation, some flat carbon steel suppliers have indicated, in particular with regard to HR, that they are only able to supply regionally within the EEA from a given steel plant. In general, suppliers mention factors such as transport costs and lead times as limitations on their supply ranges. The limiting nature of transport costs appears to apply particularly to products for which transport cost represents a higher share of the value of the product, such as commodity HR, whereas higher value specialty products can travel farther. (246) A flat carbon steel supplier explains in this respect: ‘Usually commodity grades are not sold that far away from the production site as that would not be economically feasible due to cost of transportation.’ (247)

(358)    Sixth, the sales share of non-Southern competitors of the Parties in Italy and in Southern Europe remained limited and relatively stable over the period 2011–2016,  as illustrated in Table 19. In Italy, the share of  non-Southern  producers  was  between [5-10]–[10-20]% and in Southern Europe the share was between [10-20]%.

7.2.5.6. Customers tend to source regionally

(359)    Feedback from customers during the market investigation appears consistent with the finding that (at least commodity) flat carbon steel is usually supplied within the same or neighbouring countries of the production sites: Nearly half of distributor  customers responding to the market investigation indicated that they sourced flat carbon steels from suppliers located in the same or nearby countries within the EEA. The rest explained that they source at least some volumes from further away (for instance across the EEA); however, only a minority explained that they source from outside the EEA. The replies were largely similar across different types of flat carbon steel products, including HR, CR and GS. (249) Replies from end-users of flat carbon steel (that is industrial customers) show even a slightly higher incidence of proximity to the supplier with approximately half or more of respondents indicating that they source from suppliers located in the same or nearby country within the EEA. Again, the replies are largely similar regardless of the type of flat carbon steel at issue. (250)

(360)    Customers have further explained in the market investigation that their sourcing from nearby production sites may be due to both their own preferences as well as to the suppliers’ willingness to supply. To that effect, a customer submits that the ‘location of the steel manufacturing plant is very important and that the proximity of the steel manufacturer to [name of the customer]'s plants is preferred’ (251) while another  market participant explains: ‘Sourcing from northern Europe by [name of the customer] is limited. Such sourcing is in general hampered by logistic costs (particularly if transport needs to go on land) and the lack of available HR volumes. For instance, ThyssenKrupp, VoestAlpine and SSAB offer no real volumes limiting their feeding to the captive market. While Tata Steel Europe has some available volumes, it sells them in northern Europe and not in Italy.’ (252) A third customer concurs: 'Northern European suppliers are not interested in supplying Southern Europe with anything but special grades. For instance, SSAB - - only  supplies special grades in Southern Europe even if it supplies commodity grades in the north. The situation is similar with regard to the German Thyssenkrupp and Salzgitter - - Sourcing commodity steel from northern to southern Europe is limited due to several factors, including: (i) higher transportation costs and logistic issues - - (ii) lack of much spare capacity – most of the steel is already committed to traditional customers in the northern market, and (iii) a competitive and fragmented Italian market which seems to be too complicated for many suppliers.’ (253)

7.2.5.7.       Prices are differentiated geographically

(361)    The Commission observes that the differences in the structure of supply, demand and non-homogeneous conditions of competition between Northern Europe and Southern Europe appear to be reflected in different prevailing prices for flat carbon steel products in the two regions.

(362)    First, the price difference between Northern and Southern Europe for HR has been acknowledged by the Notifying Party. (254) The price difference has also been confirmed in the market investigation by both competitors and customers. (255) Most of customers as well as competitors who have taken a position have further indicated that there is a difference in competitive conditions, many of them explicitly  indicating a price difference, between Southern and Northern Europe, with regard to CR and galvanised (HDG and EG) flat carbon steel as well. (256)  The price  differences between Northern Europe and Southern Europe are depicted in Figure 45, Figure 46, and Figure 47 for HR, CR, and HDG, respectively. (257)

Figure 45 – HR price in Northern Europe and Southern Europe (258)

[Parties’ submissions]

Figure 46 – CR price in Northern Europe and Southern Europe (259)

[Parties’ submissions]

Figure 47 – HDG price in Northern Europe and Southern Europe (260)

[Parties’ submissions]

(363)    Second, the Commission considers that the price difference between Northern and Southern Europe cannot be solely explained by differences in transport costs for non-EEA exporters. This is because the difference in terms of seaborne transport costs for exporters between reaching Southern Europe (Genoa, Italy) and reaching Northern Europe (Antwerp, Belgium) from the main non-EEA exporting countries to the EEA is significantly lower than the difference in prices between the two regions (~20 EUR/t). The transport costs are presented in Table 20.

(364)    Third, price differences also apply to like for like products. Figure 48 Figure 49 and Figure 50 depict the price, on an ex-works basis, of ArcelorMittal's HR grades S235JR, CR grade DC01 and HDG grade DX51D+Z, respectively, in a selection   of EEA countries over time. [Commercial information]. Thus, the difference in the average prices between those Southern and Non-Southern European countries  appears not to be explainable by differences in product mixes in the two regions (for instance more commodity products in the south and more specialty products in the north).

Figure 48 – [Commercial information] (261)

[Commercial information]

Figure 49 – [Commercial information] (262)

[Commercial information]

Figure 50 – [Commercial information] (263)

[Commercial information]

(365)    Fourth, the Commission considers that the price correlation analysis provided by the Notifying Party does not call into question the price differences and market differentiation.

(366)    The Commission notes that, as explained in recital (236), in general, a price correlation analysis can only provide indirect evidence of market definition, and is more suitable as a ‘separation’ test rather than an ‘inclusion’ test. That is to say, it can provide insights on whether two products/regions do not belong to the same relevant market, rather than whether they belong to the same relevant market. This is because the evidentiary value of a price correlation analysis is highly dependent on controlling for common cost and demand factors. (264)

(367)    For instance in carbon steel, and as acknowledged by ArcelorMittal, prices for raw materials such as iron ore and coking coal affect the prices of carbon steel and they may thus constitute common cost factors across different regions even if those regions were not linked. Price level analysis does not control for such factors.

(368)    Moreover, a version of the correlation submitted by the Notifying Party that attempts to control for common factors affecting the price series in different countries in the EEA (265) finds that the correlation between Italy and Spain is significantly higher than between Italy or Spain and any other countries in the EEA. (266) According to the Notifying Party's own reasoning, this finding would point towards the conditions of competition  in  Italy and  Spain  –  which  constitute  […]%  of  all  flat  carbon steel demand (HR, CR, HDG, EG) in the Southern European region as discussed  in Section 7.2.5.1 – being more similar to each other than to other EEA countries.

(369)    All in all, the Commission does not regard the analysis submitted by the Notifying Party as sufficiently informative on its own as regards the existence and the extent of price arbitrage between the Northern and Southern European regions. While not providing conclusive evidence on Southern Europe being a distinct market, the Notifying Party's correlation study, when accounting to some extent for common factors that may drive prices across the EEA, suggests a degree of differentiation in the competitive conditions across EEA countries, with conditions more similar in Italy and Spain that are also the main steel demand hubs in the Southern European region.

(370)    Fifth, respondents to the market investigation appear to point at the differences in structures of demand and supply across Northern and Southern Europe as an explanation for the difference in prices. A flat carbon steel producer explains: ‘In Southern Europe, where the customer base is very fragmented, SSCs tend to stock more materials and speculate on price changes by buying in bulk on the spot’. (267) Another respondent more expressly links the price difference between the north and the south of Europe to a different balance of demand and supply: ‘[T]he price difference may be the outcome of the historic imbalance between supply and demand of HR in different regions of Europe. For instance, there is less local supply  available in Italy and Spain and thus more room for imports. At the same time, a number of harbours located in southern Europe, the logistics to southern Europe from most exporting countries are easier and the sea route shorter.’ (268)

7.2.5.8.       ArcelorMittal appears to pursue a regional, or even country-based, sales strategy

(371)    [Commercial information].

Figure 51 – ArcelorMittal's flat carbon steel divisions in the EEA

[…]

(372)    [Internal document].

Figure 52 – ArcelorMittal internal document (269)

[…]

Figure 53 – ArcelorMittal internal document (270)

[…]

Figure 54 – ArcelorMittal internal document (271)

[…]

(373)    The Commission considers that the fact that ArcelorMittal appears to have a differentiated sales strategy within the EEA is in itself not sufficient to define separate markets within the EEA; however, it supports the finding that the EEA markets are at least differentiated.

7.2.5.9.Transport costs within the EEA

(374)    The Commission has taken into account the factors regarding the transport costs for Northern European producers to reach Northern Italy presented by the Notifying Party in support of an EEA-wide scope of the geographic market for flat carbon steel products (see recitals (316) to (318)). At the same time, the Commission considers that those factors do not call into question the differentiation of the competitive conditions between different regions in the EEA and in particular between Southern Europe and more northern parts of the EEA.

(375)    As explained in Sections 7.2.5.4 and 7.2.5.5, the Commission does not contest that Northern European producers supply to Southern Europe, including  to  Northern Italy. This suggests that the transport costs, although contributing to the geographic differentiation in the EEA, do not represent a major constraint for plants located outside Southern Europe to supply that region. However, those producers generally supply limited quantities to Southern Europe, and in particular to Italy, compared to both their own production and the market size in  the region  (see Sections  7.2.5.5 and 9.3.2). This suggests that the geographic differentiation in the EEA is more related to the incentives of producers to supply certain areas rather than their ability.

(376)    On the basis of the data submitted by ArcelorMittal and Ilva, the Commission computed the average transport costs to serve Northern Italy, as shown in Figure 55. Those transport costs differs somewhat from those presented by the Notifying Party in Figure 30, as the Notifying Party's figures are based solely on road transport, to fictitious  customer   locations   whereas   the   Commission's   figures   reflect   actual transport costs for real delivery (that accounts for the actual logistics utilised, for instance sea transport, for the actual locations of the Parties' customers and the actual customers' delivery needs). [Parties' submissions].

(377)     […].

Figure 55 - Transport cost for ArcelorMittal and Ilva to North of Italy

[…]

Source: Commission's calculation based on data submitted by Ilva and ArcelorMittal in response to RFI 26

(378)    This shows that transport costs introduce an element of differentiation that, all else equal, will depend on the distance between the plant and the customers, as well as on the means of transport available. Moreover, such differentiation will be more pronounced for products that command a lower margin like HR, and less pronounced for products that command a higher margin such as HDG.

(379)    The Commission has further conducted a market investigation on transport costs.  The investigation shows that transport cost to supply to  Italy vary from  45 EUR/t   to 100 EUR/t depending on the location of the supplier. (272) The evidence collected also shows that transport costs vary both by region and to a certain extent also across suppliers located in the same region (although this latter difference might be more related to different estimation methodologies applied by different suppliers) (273). Taking into account the responses to its market investigation the Commission considers that the actual transaction level data of ArcelorMittal and Ilva are more illustrative to describe the transport cost differentiation as they refer to actual transactions with customers located in the areas of interest (in particular North of Italy).

(380)    All in all, the Commission considers that differences in transport costs alone do not explain the differentiated pattern of sales that are observed across the EEA. Rather, the different pattern of sales are explained by a combination of factors that, beyond transport costs, include also the supply and demand characteristics of the different sub-regions of the EEA.

7.2.6. Exclusion of the Nordic countries from an EEA-wide market would not materially affect the outcome of the competitive assessment

(381)    The Commission recalls that, in SSAB / Rautaruukki, the Commission considered  that there was at least a serious possibility that three Nordic countries (Finland, Sweden and Norway) constituted a distinct geographic market at least for (commodity) HR, CR and OC. (274)

(382)    The Parties have no overlaps in those three Nordic countries as Ilva makes no supplies there. ArcelorMittal’s merchant market shares in those Nordic countries remain below 20% for HR and CR and below 30% in GS. Neither of the Parties have any significant production assets in those Nordic countries.

(383)    Moreover, the Commission observes that the overall flat carbon steel market in those Nordic countries is limited compared to that of the rest of the EEA. For instance, based on the information provided by the Notifying Party, the total merchant market volumes of the Nordic countries combined constituted less than […]% of the total EEA volumes for each of HR, CR and GS.

(384)    Therefore, while the Parties’ market position might be slightly stronger in a potential ‘EEA less Nordics’ market compared to a potential EEA-wide market, the difference does not appear to be of such magnitude as to affect the outcome of the competitive assessment in the present case. It is thus not necessary to discuss further whether or not the Nordic countries are part of a potential (differentiated) EEA-wide market.

7.2.7. Conclusion on the geographic market for the production and supply of flat carbon steel products

(385)    On balance and considering all evidence available to it, the Commission finds that  the relevant markets for the production and supply of flat carbon steel products  where the Transaction raises competitive concerns (HR, CR, HDG and EG) are EEA-wide  with,  however,  strong  indications  of  geographic  differentiation.  Such geographic differentiation is particularly pronounced for lower value added products, such as HR, whereas it  is less pronounced for high value added products such as HDG.

(386)    With regard to the Nordics, it can be left open whether or not they are part of  the EEA-wide markets as discussed in Section 7.2.6.

(387)    The Commission will take account of the geographic differentiation across the EEA, as evidenced by the elements discussed in this Section, in its competitive assessment. The Commission will focus its analysis on the competitive conditions in the parts of the relevant market where the competitive interaction between the Parties is strongest and where the most significant effects of the Transaction are likely to be felt (that is Southern Europe, which is largely driven by Italy).

(388)    Regarding the relevant geographic market definition for metallic coated steel for packaging, the considerations that apply to HR, CR, HDG and EG also appear to be to a large extent relevant, However, the Commission finds that the precise  geographic market definition for metallic coated steel for packaging can be left open for the purposes of assessing the Transaction, as the Transaction does not significantly impede effective competition for these products regardless of the  market definition. This also applies to OC where only ArcelorMittal is active.

(389)    Finally, while a definition of a geographic market broader than the EEA is not warranted in this case, the Commission has considered any competitive constraint exerted by non-EEA imports in the context of the competitive assessment of the Transaction, as further described in Sections 5.3.3, 9.4.5, 9.5.5, and 9.7.5.

7.3. Production and supply of welded carbon steel tubes

7.3.1. Small welded carbon steel tubes

7.3.1.1. Previous cases

(390)    The Commission has previously left it open whether the relevant market for the production and supply of carbon steel tubes was EEA-wide. (275)

7.3.1.2. The Notifying Party's views

(391)    The Notifying Party submits that the relevant market for small welded carbon steel tubes is EEA-wide, in particular due to the absence of barriers to EEA trade and given the price convergence throughout the EEA.

7.3.1.3. The Commission's assessment

(392)    The response to the market investigation was split between customers and competitors on the question of geographic market for small welded carbon tubes, although there are indications that there are price differences between Eastern Europe and Western Europe. (276)

(393)    The Commission finds however that for the purposes of the present case, it is not necessary to conclude on the exact scope of the geographic market for small welded carbon steel tubes, as the Transaction would not significantly impede effective competition on this market, however defined.

7.3.2. Large welded carbon steel tubes

7.3.2.1. Previous cases

(394)    The Commission has previously considered whether the relevant geographic market for the production and supply of any sized steel tubes is EEA-wide or wider, but ultimately left the market definition open. (277)

7.3.2.2. The Notifying Party's views

(395)    The Notifying Party submits that the geographic market should be at least EEA-wide, if not worldwide. In this regard, the Notifying Party refers to the Europipes case where the Commission found that demand for large longitudinally welded and spiral welded tubes is driven largely by the large projects of the international oil and gas companies. (278) In such cases, the bulk of sales are for very large tonnages with no particular geographical or temporal distribution system. The usual procedure for the purchase of these tubes is by international procurement. Whereas the Commission found in that case that there are factors suggesting that the relevant geographic  market for large longitudinally welded and spiral welded tubes is worldwide, it considered that the market would, in any event, not be smaller than 'Western Europe'.

7.3.2.3. The Commission's assessment

(396)    The Commission finds that for the purposes of the present case, it is not necessary to conclude on the exact scope of the geographic market for large welded carbon steel tubes, as the Transaction would not significantly impede effective competition on  this market, however defined.

7.4. Distribution of flat carbon steel

(397)    The Commission has in previous cases found the geographic markets for distribution of flat carbon steel to be national or at most cross-border regional, in light  of transport costs and customers' lead time requirements (regardless of whether through SSCs, SCs or oxy-cutting centres). (279)

(398)    The Notifying Party agrees with the  approach  taken  in  Commission  precedents. No information came to the knowledge of the Commission during the market investigation which would have called Commission precedents into question either.

(399)    Therefore, the Commission considers that, for the purposes of the Transaction the markets for the distribution of flat carbon steel products are national in scope regardless of the type of distribution facility.

8.  COMPETITIVE ASSESSMENT PRINCIPLES

(400)    According to the Commission's Guidelines on the assessment of horizontal mergers under the Council Regulation on the control of concentrations between undertakings (the 'Horizontal Merger Guidelines') (280) there are two main ways in which horizontal mergers may significantly impede effective competition:

(a)   by eliminating important competitive constraints on one or more firms, which consequently would have increased market power, without resorting to coordinated behaviour (non-coordinated effects) – the non-coordinated effects of the Transaction will be assessed in Section 9; and/or

(b)   by changing the nature of competition in such a way that firms that previously were not coordinating their behaviour are now significantly more likely to coordinate and raise prices or otherwise harm effective competition. A merger may also make coordination easier, more stable or more effective for firms which were coordinating prior to the merger (coordinated effects) (281) – the coordinated effects of the Transaction will be assessed in Section 10.

8.1. Framework of the assessment

8.1.1. Application of the priority rule

(401)    The proposed concentration takes place against the background of consolidation in the European steel industry, which may also include a merger between the activities of ThyssenKrupp and Tata Steel, (282) respectively the second and third largest EEA steelmaker.

(402)    The situation where concentrations affecting the same market are notified within a short time frame from each other is an unusual, though not an exceptional one. In previous cases, the Commission decided such cases by applying a priority principle based on the date of notification. (283) Although there is no explicit provision in the Merger Regulation relating to the appropriate treatment of concentrations which are notified almost simultaneously, the Commission considers that the priority principle is inherent in the system of the Merger Regulation. Article 6(1) provides for the Commission to ‘examine the notification as soon as it is received’ and which sets time limits by reference to the date of notification. The date of notification is therefore the only criterion that can ensure sufficient legal certainty, transparency and objectivity and respect the other provisions and aims of the Merger Regulation.

(403)    Consistent with its previous practice, the Commission assessed the proposed concentration according to a priority principle (‘first come, first served’ approach) based on the date of its notification. The starting point of the Commission's assessment of the Transaction is therefore a market structure in which Tata Steel and ThyssenKrupp act as independent EEA steel producers.

8.1.2.  Determination of the relevant counterfactual – failing firm defence

8.1.2.1.  The Notifying Party's views

(404)    In its submissions, the Notifying Party argued a failing firm defence (‘FFD’), claiming that the conditions for the application of the FFD are met.

(405)    The main reasons put forward by the Notifying Party were that: (i) Ilva is not a credible and sustainable supplier as it is facing industrial, commercial and environmental problems due to historic underinvestment; (ii) Ilva is structurally loss- making, quickly running out of money and facing bankruptcy liquidation  under Italian insolvency law; (iii) no viable alternative buyer to ArcelorMittal exists as the alternative bidding consortium has been unwound and its industrial plan was  flawed; (iv) absent a clearance of the Transaction Ilva would lose its environmental permit and inevitably exit the market.

(406)    The Notifying Party also submitted that, if the Commission were to reject the FFD, it should at least take those elements into account in the determination of the relevant counterfactual for the competitive assessment of the Transaction.

(407)    Finally, the Notifying Party argued that the status quo would be the most optimistic counterfactual scenario. In this vein the Commission should not take into account the full capacity of Ilva but only the capacity that can be operated under the current environmental restrictions. In other words, the Commission should not assume that Ilva would return to its historical peak supply levels absent the Transaction.

(408)    While not taking a position on the issue of the counterfactual scenario, Ilva provided arguments regarding its own situation, stressing that it is under extraordinary administration, that it is facing commercial and financial difficulties, that its  activities cannot continue due to major environmental constraints, that a second timely and successful tender is highly unrealistic, and that the failure to finalise the sale would entail dramatic social and environmental consequences. Ilva further submitted that the proposed Transaction would be the only viable solution to rescue Ilva in a timely manner and avoid the significant loss of capacity on the EEA market.

8.1.2.2. Legal framework

(409)    Assessing the competitive effects of a proposed concentration under the Merger Regulation involves a comparison of the competitive conditions that would result from the notified merger with the conditions that would have prevailed in the  absence of the merger. In principle, the competitive conditions existing at the time of notification  constitute  the  relevant  framework of comparison for evaluating the effects of a concentration. (284) However, in some circumstances the Commission may take into account future changes to the market that can reasonably be predicted. (285)

(410)    One of the situations in which an alternative counterfactual may be relevant is the potential exit of a firm from a market.

(411)    As regards this specific situation, according to the Horizontal Merger Guidelines and the relevant case law, the Commission may decide that an otherwise problematic merger is nevertheless compatible with the internal market if one of the merging parties is a failing firm. The basic requirement is that the deterioration of the competitive structure that follows the merger cannot be said to be caused by the merger. This requirement is met when the competitive structure of the market would deteriorate to at least the same extent in the absence of the merger.

(412)    According to the Horizontal Merger Guidelines and the relevant case law, three criteria are especially relevant in determining whether an entity is to be regarded as a ‘failing firm’. First, the allegedly failing firm would in the near future be forced out of the market because of financial difficulties if not taken over by another undertaking (first criterion of the failing firm analysis). Second, there is no less anti- competitive alternative purchase than the notified merger (second criterion of the failing firm analysis). Third, in the absence of a merger, the assets of the failing firm would inevitably exit the market (third criterion of the failing firm analysis). The guidelines clarify that the inevitability of the assets of the failing firm leaving the market may underlie a finding that the market share of the failing firm would in any event accrue to the other merging party.

(413)    It is for the Parties to provide all the relevant information necessary to demonstrate that these criteria are met and thus that the expected deterioration of the competitive structure that follows the merger would not be caused by the merger.

8.1.2.3. The Commission's assessment: failing firm defence and exiting firm counterfactual

(414)    The Commission considers that the first and second contentions of the Notifying Party, that is (i) that the conditions for a FFD are met, and (ii) should the  Commission reject a FFD it should nonetheless take the circumstances and the risk that the firm exit the market into account in the determination of the counterfactual scenario, have in essence the same object.

(415)    Should the Commission take into account market exit as a relevant counterfactual scenario in order to assess the effects of the merger, it would in essence be tantamount to the acceptance of a FFD.

(416)    After an analysis of the arguments submitted by the Parties, the Commission does  not consider that the conditions for invoking the FFD are met, since at least the second and the third criterion of the failing firm analysis are not fulfilled.

(417)    As to the second criterion, the Commission considers that the Notifying Party has not demonstrated such criterion to be met.

(418)    First, as to the absence of alternative buyers, the Commission observes that the assets were awarded as a result of a competitive procedure where two bidders submitted bids. Absent the Transaction, the most likely counterfactual scenario at the time of the merger would thus have been that another buyer would have been awarded Ilva  as a going concern.

(419)    Indeed, the Ilva assets were awarded at the conclusion of a competitive bidding process, where another consortium (Acciaitalia) submitted a binding offer. Acciaitalia's industrial and environmental plans were assessed and scored by the Italian authorities. ArcelorMittal's offer was eventually retained because it offered a higher consideration for the Ilva assets, in view of the fact that the financial consideration was given a considerable weighing in deciding the award of the bid.

(420)    Second, the Notifying Party further submitted that the Acciaitalia consortium was dissolved, with two of the members of the consortium, Arvedi and Cassa Depositi e Prestiti, having parted from the Acciaitalia consortium. This argument does not affect the Commission’s analysis.

(421)    In the first place, the alterations in the composition of the alternative bidding consortium do not affect the circumstance that ArcelorMittal's bid was part of a competitive bidding process, absent which an alternative buyer would likely have purchased the assets.

(422)    In the second place, after the dissolution and upon the announcement of the bid award, the remaining two members of Acciaitalia, Indian steelmaker Jindal  and Italian group Delfin (owned by the Del Vecchio family) confirmed their interest and submitted a binding offer for a higher consideration than that retained in the ArcelorMittal bid. Even though the latter offer was not admitted – as the Italian government indicated it could not re-open the bidding process at the time of adjudication – this further supports the view that, absent the Transaction, at the time of notification Ilva would likely have been acquired by another bidder as a going concern.

(423)    Third, as to the Notifying Party's allegation that the delays incurred and the degradation of Ilva's assets in the meantime would make it unlikely that the alternative bidder would be still in a position of acquiring Ilva, the Commission observes that the alleged further degradation, as well as the alleged further deterioration of the competitive structure, cannot be taken into account.

(424)    The Commission considers that the competitive situation existing at the time of the merger (286) is the relevant counterfactual for establishing whether the  Transaction raises concerns. In the context of a bid, the assessment of whether an alternative to the notified transaction would exist in a counterfactual scenario should be justified  by objective and verifiable criteria at the time the choice between alternative bids is made. This assessment cannot be made contingent on a deterioration of the prospects of alternative transactions that happen after, or even because, the notified  transaction has been chosen. The Commission is bound to assess potential anti-competitive effects stemming from a transaction, and the sellers of an asset may decide to take this into account when making a choice between alternative bidders. When the transaction chosen by the parties leads to a deterioration of the market structure, and to the extent that alternative options are available at the time of such choice, the parties knowingly take a risk. The risk taken with the choice of a potentially anti- competitive choice cannot be shifted on the Commission in that the Commission cannot refrain from a full assessment of the transaction's potential anti-competitive effects. Concluding otherwise would deprive the very assessment the Commission is bound to carry out under the Merger Regulation of its purpose.

(425)    As to the third criterion, the Commission considers that the Notifying Party has not demonstrated that, absent the Transaction, the assets would inevitably exit the  market.

(426)    First, the Commission's State Aid decision concerning Ilva acknowledges that the value of the asset, as opposed to the company under administration Ilva A.S., remained very high, notwithstanding the economic difficulties as well as the current liabilities.

(427)    Second, the value and potential of the assets is reflected by the offer of ArcelorMittal itself. […], (287)  […] (288) […].

(428)    Third, concerning Ilva's HR coil production efficiency, [commercial information] (289). […]. (290)

(429)    Fourth, apart from the monetary valuation, it is to be emphasised that Ilva remains  the largest integrated steelmaking site in Southern Europe, strategically located within Europe's second largest steel consuming country and at the centre of the demand for HR. (291)

(430)    The strategic value of the assets is also acknowledged in ArcelorMittal's internal documents, […]. (292)

(431)    Moreover, the Commission finds that, as regards the inevitability of the assets of the failing firm leaving the market, the Notifying Party has not provided any evidence that the market share of the failing firm would in any event accrue to the other merging party.

(432)    The Commission therefore finds that the Notifying Party failed to demonstrate to the requisite standard that the conditions for a FFD are met.

(433)    As explained in this Section, in assessing the counterfactual proposed by the Notifying Party, (that is the likelihood of Ilva exiting the market), the Commission applies  the  same  three  criteria  of  the  FFD.  Hence,  the  Commission  rejects  the submission of the Notifying Party and considers that the relevant counterfactual scenario is the one prevailing at the time of the merger.

8.1.2.4.  The Commission's assessment: the relevant counterfactual scenario

(434)    The Notifying Party argued that the status quo, that is the current production levels and the capacity levels that can be currently utilised, should be the most optimistic counterfactual for the assessment.

(435)    In principle, as discussed in recitals (9) and (115) to (117), due to the fact that Ilva's assets were awarded as a result of a competitive bidding process, and thus the most likely alternative scenario to the Transaction would have been an alternative acquisition, the Commission considers that its assessment should not disregard the full capacity of Ilva, as this capacity exists irrespective of the fact that it can  currently not be operated. This is also supported by the fact that even the alternative bidder would have been obliged to perform the required environmental investments. (293)  [Internal document]. (294)

(436)    However, while alternative capacity figures will be provided where appropriate, the Commission has based its assessment on a conservative approach that only accounts the capacity and production of the assets as currently operated, even if such an approach may underestimate the likely effects of the Transaction.

8.1.2.5. The Commission's assessment: the planned restoration of capacity by ArcelorMittal

(437)    The Notifying Party also argued that, because under its industrial plan it plans to complete the environmental investment and lift the capacity cap, the Commission’s assessment should take this into account as a pro-competitive effect of the Transaction, as it would increase market supply in the EEA, and in particular in Italy.

(438)    The Commission finds that the argument of the Notifying Party is misplaced.

(439)    First, as discussed in the previous Section 8.1.2.5, as the capacity of Ilva exists and environmental remediation was also required by the alternative buyer, the restoration of Ilva capacity should at best be considered as neutral with respect to the counterfactual scenario.

(440)    Second, the restoration of capacity by the merged entity is not predictive of the impact of the Transaction on prices. In particular, while Ilva's capacity is currently operated independently of ArcelorMittal, and in competition with it, as a result of the Transaction the total capacity would be controlled by the merged entity, thus adding no competitive pressure on the merged entity itself.

(441)    Third, on a proper construction, since large capacities may be indicative of market power, a larger addition of capacity to the merged entity than that currently operated by Ilva does not by itself remove potential concerns as to the concentration of such capacity in the first place. The additional volumes in the hands of the Parties would further concentrate capacity where the Notifying Party previously competed with Ilva.

(442)    Moreover, as regards possible capacity expansions by the merged entity post- Transaction, this is not necessarily predictive of total volumes that may be supplied into the relevant markets as to justify a pro-competitive effect brought about by the Transaction. Market demand in Europe is increasing and could absorb further production over the longer time period over which […] (295) […]. (296) Controlling capacity that was previously independently owned, ArcelorMittal could have the incentive and ability to adopt an offsetting strategy if its available supply were to depress prices, for instance in reaction to a drop in demand. This is because, in contrast with the scenario where the capacity would be independently operated, ArcelorMittal would internalise the effect of those sales.

(443)    In such a situation, irrespective of its industrial plan, ArcelorMittal may decide to direct quantities elsewhere, including by diverting increased volumes to export markets outside the EEA, or even by reducing capacity utilisation and, ultimately, shutting down or reducing capacity at other steel mills.

(444)    While the incentive and ability of ArcelorMittal not to adopt strategies that may depress its sales prices will be largely discussed in the competitive assessment, the […] are demonstrated by the past strategy of ArcelorMittal. (297)

8.2. Market share methodology

(445)    Market shares and concentration levels provide useful first indications of the market structure and of the competitive importance of both the merging parties and their competitors. (298) As explained further in Section 8.2.2, the Commission considers different market share measures to be relevant indicators for market power in this case:  (i) sales,  or merchant,  market  shares  on  the basis  of sales  to  third   parties;

(ii)   shares  of  production  capacities  at  the  relevant  level  of  the  value  chain; and

(iii)    consumption shares which include third party sales and, in addition, internal consumption of the product by integrated producers.

8.2.1. The Notifying Party's views

(446)    In its Reply to the SO, the Notifying Party took issue with the Commission's use of capacity and consumption based market shares.

(447)    First, the Notifying Party argued that the SO's EEA-wide capacity shares exclude imports, whereas excluding imports from share metrics cannot provide an accurate assessment of competitive dynamics in the EEA. Supplies from outside the EEA, which account for [20-30]% of HR, [20-30]% of CR and [10-20]% of GS, exert a significant competitive constraint on the EEA flat carbon steel industry. When  import supplies are included, the resulting ‘adjusted’ capacity shares would closely resemble sales shares. The Parties combined EEA shares on that basis would be [30-40]% in HR, [30-40]% in CR, [30-40]% in GS and would hence not be  indicative of competitive concerns.

(448)    Second, the Notifying Party argued that consumption shares, as well as capacity shares, wrongly include captive production and there is no mechanism by which the existence of captive production would increase the scope for the Transaction to give rise to anti-competitive effects in merchant supply. The Notifying Party argued that the Transaction would not reduce any pre-existing scope for the Parties' captive HR and CR production to act as a constraint on HR and CR merchant market prices as captive production is not available for merchant supply and cannot be leveraged downstream to increase upstream merchant prices. The inclusion of captive production in market shares metrics would hence be inappropriate and would significantly overstate the Transaction's competitive impact. Moreover, the SO's consumption shares would fail to take account of downstream imports. According to the Notifying Party, when correcting for this inconsistency, the combined EEA-wide consumption share moves from [30-40]% to [30-40]% for HR and from [40-50]% to [30-40]% for CR. (299).

(449)    Third, the Notifying Party argued that the Parties' rivals would have the ability and incentive to defeat any price increase by the merged entity in HR or CR. To take account of available rival spare capacity, the Notifying Party submitted ‘merchant capacity shares’ as an alternative share metric. Merchant capacity shares are estimated by adding to the firms' merchant sales in the EEA their spare capacity in a given product segment (either HR or CR). Taking account of Ilva's upstream blast furnace bottleneck, which would effectively limit its HR capacity, the Notifying  Party noted that these market shares are similar to their merchant sales shares. In HR, the Parties' merchant capacity share is [20-30]% (compared to merchant sales share  of [20-30]%) and in CR the combined merchant capacity share is [30-40]% (as compared to a merchant sales share of [30-40]%). The merchant capacity shares of Tata Steel and ThyssenKrupp are significantly higher than their merchant sales shares, highlighting the ability of these rivals to bring sizable volumes to the market within a reasonable timeframe and hence to constrain the merged entity.  Furthermore, according to the Notifying Party, the merchant capacity shares of imports, which are computed solely on the basis of sales, ignore a big volume of global divertible capacity.

(450)    In conclusion, the Notifying Party asserted that when properly assessed (on the basis of supply shares, adjusted capacity shares or merchant capacity shares), the Parties' combined EEA-wide market shares would indicate that the Transaction cannot raise any concerns, particularly as regards HR and CR.

8.2.2.              The Commission's assessment

(451)    The choice of market share measure depends on the circumstances of the specific industry in question. Different market share measures may have different advantages and shortcomings in indicating market power. It may therefore be useful to analyse a combination of different market share measure as complementary first indicators for market power.

(452)    The structure of the flat carbon steel industry imposes some considerations as to what types of market shares are relevant first indicators for the assessment of market power, in the sense of the Horizontal Merger Guidelines.

(453)    First, merchant sales shares provide a measure of the relative positions of the different producers, including importers, as a supplier to third parties. Merchant sales shares are also informative to assess the geographic differentiation of firms' sales across the EEA. Indeed, in contrast to capacity and production which is usually concentrated in relatively few locations, the pattern of merchant sales are more widespread  across  a  number  of  different  countries.  Merchant  sales  shares  in a specific geographic region can therefore provide useful information on the ability, readiness and incentive to serve specific geographic areas.

(454)    Second, competition in the flat carbon steel industry is largely driven by primary steelmaking capabilities, that is crude steel capacity, which is closely linked and can be approximated to HR capacity. In particular, the largest investment required for a hypothetical greenfield integrated flat carbon steel plant, and hence the highest barrier to entry, relates to primary steel production assets. The number of rolling  lines is then generally calibrated based on the crude steel production capability of the plant, and the investment required to build a new line is small compared to the investment required for a blast furnace or a BOF. Hence, primary steel production represents a binding constraint on the overall production of a flat carbon steel producer in the EEA. (300)

(455)    Following primary steel production of slabs, flat carbon steel suppliers transform most of the slabs they produce first into HR coils. (301) A large part of HR coils is then further transformed along the flat carbon steel value chain, either by integrated steel producers themselves (through captive use of HR) or by non-integrated processors which buy HR coils on the merchant market. (302) Similarly, CR can either be further processed internally or sold to third party end customers or processors. The production capabilities of different suppliers at each level of the value chain are best captured by capacity shares.

(456)    The importance of capacity as a competitive parameter is also reflected by how steel manufacturers assess the market. […]. (303) […].

(457)    To capture the position of integrated producers in terms of production capabilities at each level of the value chain, and in line with previous decisions, (304) the Commission therefore examines capacity shares in addition to merchant sales shares.

(458)    For its computation of the capacity shares, the Commission used data provided by  the Parties' main EEA competitors, complemented by data provided by the Notifying Party. The Commission devised two scenarios for the capacity shares: a first scenario including only the capacities that are readily available, and a second scenario including the capacity that could potentially be brought to the market with limited constraints in terms of investments and business decisions (for example if HR capacity could be increased by the insourcing of slabs both internally from other plants or from third parties).

(459)    Third, for the purposes of this Decision, the Commission has computed  ‘consumption shares’ as a further indicator of market power in HR. This metric provides an indicator that reflects the importance of different producers in terms of production capabilities (including production for captive use) while also   accounting for the competitive importance of imports. It can further provide a geographically differentiated picture that does not require an arbitrary allocation of capacity to a specific use or geographic area. Consumption shares approximate the share of overall demand of a given flat carbon steel product that is fulfilled by a given entity, for both EEA and non-EEA suppliers. Consumption encompasses both the merchant market demand, as well as the captive demand of a given product by flat carbon steel producers. In the case of HR, this measure captures the position of each producer (including non-EEA producers) in supplying HR as an input for downstream use (regardless of whether this input is supplied internally within the same company or to a third party). The consumption shares of HR at EEA level are computed as the HR production (not capacity) by all EEA suppliers, minus HR export sales from the EEA suppliers, plus HR sales in the EEA by non-EEA suppliers. (305)

(460)    Consumption shares are particularly relevant for HR and CR, the flat carbon steel products that have a significant captive demand and that are transformed to further downstream products in sizeable volumes.

(461)    Internal consumption of HR coils for the production of downstream products that represents an indirect constraint to the HR sales on the merchant market, as the internal consumption can be redirected to the merchant market. Likewise, to the extent that rolling capacity allows, sales to the merchant market of upstream products (for example HR) can be channelled for further transformation downstream (for example CR and GS), hence representing an indirect constraint to downstream products. Similarly, as HR is the main input for downstream flat steel products, the price of HR affects downstream prices and hence the downstream sales of integrated steel makers, as explained in Section 5.4.

(462)    In summary, the Commission considers three complementary market share measures to be informative as first indicators of market power in the sense of the HMG:

(1)   Sales shares are useful to assess the concrete competitive interaction, also in different and differentiated parts of the relevant market, by also taking into account constraints from outside the relevant market, such as imports, which  do not have a structural presence in the EEA. Merchant sales shares, in particular, reflect the sales to third parties and the role of imports in supplying third parties. These can be computed on an EEA basis as well as in sub-regions or individual countries.

(2)   Capacity shares which provide a direct indication of production capabilities on the EEA level, which the Commission regards an important driver of competitive dynamics among EEA suppliers. The Commission considers capacity shares are the most appropriate and informative starting point, as they constitute a more structural metric of market power than sales shares based on the capabilities of EEA suppliers. 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. The importance of capacity shares as a structural metric applies in particular to HR capacity shares, which represent the overall binding constraint not only for HR, but also for the majority of the downstream flat carbon steel products, such as CR and HDG/EG. This measure captures the structure of the market at a level of production where barriers to entry are the highest, and is an indicator of the capabilities of a steelmaker also at further level of the supply chain, and as such a potential indication of its market power in line with the general features of the steel industry as described in Section 5.1.3.1 of this Decision,

(3)   Consumption shares which, by focussing on consumption of products at  a given level of the value chain in a specific geographic area (including internal consumption or captive use by integrated player and imports), provide an indicator of the role of each producer (including imports) in supplying internal and external consumption in that area.

(463)    As regards the Notifying Party's arguments, the Commission, in the first place, acknowledges that capacity shares do not capture the role of imports. The role of imports is accounted for in the Commission's alternative market share measures.

(464)    In the second place, the Commission considers that capacity and consumption shares, which include capacity or production volumes for captive use are relevant indicators for market power and for the structural change induced by the Transaction at each level of the value change for the reasons set out in recitals (455) – (457).

(465)    In contrast to the Parties arguments, the Commission considers that there are both demand side and supply side links which would be ignored by a pure focus on merchant market shares.

(466)    On the supply side, integrated producers face a trade-off between using HR (respectively CR) captively to produce downstream products and selling it to third parties. While integrated producers may prioritise integrated production as long as this generates a greater contribution to their overall profits, the existence of spare capacity in downstream markets despite higher contribution margins by an integrated producer on downstream products, indicates that, contrary to the position of the Notifying Party, integrated producers do make a trade-off between increased upstream and downstream sales at the margin. For example, there is spare capacity in the EEA market of CR, despite its higher margins compared to HR. Such spare capacity can be explained by a lack of demand for CR products, which however implies that further CR sales could only be made at the cost of reducing downstream profits (by depressing price and contribution margins on all CR sales). This strategic trade-off of an integrated supplier between upstream and downstream does not lead him to always push downstream sales with higher contribution margins. The argument that Ilva prioritised its integrated activities following its blast furnace shutdowns in 2012 does not refute this observation. Contrary to the Notifying Party's view, the combination of two integrated producers with significant captive consumption of HR and CR removes the competitive constraint between the Parties on the upstream markets not only from their actual sales on the upstream market but also from potential redirection of parts of their captive volumes.

(467)    On the demand side, in contrast to the Notifying Party's claim, a significant share of merchant market sales of HR and CR products by the Parties are not to end users but to customers that are in downstream competition with the Parties. For example,  based on 2016 sales data provided by the Parties, more than […]% of HR sales in Italy of both ArcelorMittal and Ilva are to re-rollers, distributors or tube makers, that is, to customers which are active in downstream markets in which the Parties are also active. Even at the EEA level, such customers account for    […]% of ArcelorMittal's HR sales and for a […] share of Ilva's HR sales. For CR in Italy, […]% of ArcelorMittal's sales are to distributors and primary transformation customers and […]% of Ilva's sales are to distributors and tube makers. Looking at the EEA for ArcelorMittal the share reduces to […]%. While the Notifying Party submitted that […]% of ArcelorMittal's and […]% of Ilva's HR merchant volumes in 2016 are to end users, the Notifying Party's designate only re-rollers as non-end users. Therefore the Notifying Party's figures indicated that […]% of Ilva's HR merchant market sales in 2016 respectively […]% of ArcelorMittal's HR merchant sales in 2016 are to re-rollers. […]. The fact that […] accounts for a […] of HR sales, for Ilva  in particular but also for ArcelorMittal, does not weaken the observation that such types of suppliers, and […] in particular, compete in downstream markets with the Parties and that by weakening competition in CR and GS, the Transaction might  also weaken the indirect constraint on the upstream markets as a result. (306) The Notifying Party's argument that the merged entity could not possibly raise prices […] also cannot be accepted, as (i) […], (ii) […], (iii) other EEA suppliers are unlikely to have the ability and the incentives to increase their HR capacity and supply post- Transaction, (307) (iv) […] itself appears to have limited ability to switch to imports following the anti-dumping measures. (308)

(468)    As explained further in Section 9.6, 9.7, 9.8 and 9.5, the Transaction would raise a SIEC in GS and also in CR. Price increases by the merged entity in GS (due to non- coordinated horizontal effects) would reduce the indirect constraint exercised by integrated GS producers on merchant demand for upstream products via such  demand side linkages, as this will increase the incentives to use upstream products internally. Similarly, price increases by the merged entity in CR (due to non- coordinated horizontal effects) would reduce the indirect constraint exercised by integrated CR producers on merchant HR demand.

(469)    Therefore, in contrast to pure merchant market shares, capacity or consumption shares, which include capacity or captive use of upstream products, are also useful to assess the effect on upstream markets resulting from (horizontal) concerns on downstream markets. Indeed, possible anti-competitive effects downstream have repercussions on upstream markets as they give integrated upstream producers an incentive to withdraw supplies from the upstream (merchant) market. This in turn weakens the indirect constraint exercised by the captive use of upstream products on the upstream markets.

(470)    In the third place, the Commission does not consider the Notifying Party's proposed ‘merchant capacity’ shares to be informative because any allocation of spare capacity across different products which are linked on the supply side as they correspond to different stages of the value chain remain arbitrary. This is obvious from the Notifying Party's approach in the Reply to the SO which double counts spare  capacity by allocating all spare capacity by different firms first to the HR merchant market and then again to the CR merchant market. Clearly spare capacity by rivals cannot simultaneously exercise a full constraint at all market levels. Moreover,  even the allocation proposed by the Notifying Party's subsequent submission (309) remains based on an arbitrary allocation of spare capacities. The Commission considers that the simplest way to avoid such arbitrary allocations is by looking at overall HR capacity shares or consumption shares, that is two of the three measures used by the Commission.

(471)    Finally, as regards the Notifying Party's argument that consumption shares for HR (respectively CR) should not only include imports of HR (respectively CR) products but also imports of further downstream products, the Commission considers that limiting the approach to imports at the respective level of the value chain is justified by the fact such imports are used to supply the consumption of the product at that level of the value chain, while imports further downstream do not feed such consumption. The Commission recognises, however, that imports further downstream might provide an indirect constraint on upstream consumption. For this reason, the Commission also presents, in recitals (496) and (502),  consumption shares that include imports further downstream in the assessment as a robustness check.

9. COMPETITIVE ASSESSMENT: NON-COORDINATED EFFECTS

9.1. Legal Framework

(472)    A merger may significantly impede effective competition in a market by eliminating important competitive constraints on one or more firms, which consequently would have increased market power, without resorting to coordinated behaviour (non- coordinated effects). (310)

(473)    Generally, a merger giving rise to such non-coordinated effects would significantly impede effective competition by creating or strengthening the dominant position of a single firm, one which, typically, would have an appreciably larger market share than the next competitor post-merger. (311)

(474)    According to well-established case law, very large market shares – 50 % or more – may in themselves be evidence of the existence of a dominant market position. The overall concentration level in a market may also provide useful information about the competitive situation.

(475)    Furthermore, non-merging firms in a given market can benefit from the reduction of competitive pressure that can result from a merger, since any price increase by merging firms may switch some demand to rival firms, which, in turn, may find it profitable to increase their prices. (312)

(476)    Mergers in oligopolistic markets involving the elimination of important competitive constraints that the parties previously exerted upon each other together with a reduction of competitive pressure on the remaining competitors may, even where there is little likelihood of coordination between the members of the oligopoly, also result in a significant impediment to competition. The Merger Regulation clarifies that all mergers giving rise to such non-coordinated effects shall also be declared incompatible with the internal market. (313)

(477)    The larger the market share, the more likely a firm is to possess market power. Moreover, the larger the addition of market share, the more likely it is that a merger will lead to a significant increase in market power. The larger the increase in the  sales base on which to enjoy higher margins after a price increase, the more likely it is that the merging firms will find such a price increase profitable despite the accompanying reduction in output. (314)

(478)    In evaluating the likelihood of non-coordinated effects potentially caused by a merger, it is important to assess to which extent the products of one merging party  are close substitutes to the products sold by the other merging party and by rival firms. The merging firms' incentive to raise prices is more likely to be constrained when rival firms produce close substitutes to the products of the merging firms than when they offer less close substitutes. (315)

(479)    Customers may have difficulties in switching to other suppliers where only few alternative suppliers exist or where the customers would face substantial switching costs. Such customers are particularly vulnerable to price increases. In particular, this may be the case for customers that have used dual sourcing from the two merging firms as a means of obtaining competitive prices. (316)

(480)    The Commission is unlikely to find that the merger will create or strengthen a dominant position or otherwise significantly impede effective competition when rival firms have available capacity and find it profitable to expand output sufficiently (emphasis added). In other words, the extent to which competitors to the merged entity constrain the merged entity from raising prices not only depends on the level  of their spare capacity but also on whether these firms have the incentive to react aggressively to a post-merger price increase. (317)

(481)    Some proposed mergers would, if allowed to proceed, significantly impede effective competition by leaving the merged firm in a position where it would have the ability and incentive to make the expansion of smaller firms and potential competitors more difficult or otherwise restrict the ability of rival firms to compete. (318)

(482)    Potential competition must also be taken into account in any competitive assessment. A merger with a potential competitor may also lead to negative competition effects where (i) the potential competitor already exerts a significant constraining influence or where there is a significant likelihood (for example plans to enter a market in a significant way) that it would grow into an effective competitive force and (ii) there is an insufficient number of other potential competitors, which could maintain sufficient competitive pressure after the merger. (319)

(483)    Countervailing buyer power is possessed by a customer if it is of a certain size, it is  of great commercial significance to the supplier and it has the ability to switch to alternative suppliers. Moreover, if only a particular segment of customers were to hold such countervailing buyer power, this may not be sufficient to offset the adverse effects of the merger. Finally, it is not enough that sufficient countervailing buyer power may exist pre-merger, it must continue to exist post-merger (removing a credible alternative supplier may reduce buyer power). (320)

(484)    For entry to be considered a sufficient competitive constraint on the merging parties, it must be shown to be likely, timely and sufficient to deter or defeat any potential anti-competitive effects of the merger. (321)

9.2. Affected markets

(485)    The Parties’ activities result on horizontally affected markets with respect to the production and supply of various finished flat carbon steel products: (i) HR, (ii)  CR,

(iv)    HDG, (v) EG and (vi) metallic coated steel for packaging (including both TP and ECCS). In addition, horizontally affected markets arise in the downstream markets of

(vii) distribution of flat carbon steel and (viii) production and supply of small welded carbon steel tubes.

(486)    In contrast, the Parties activities in semi-finished flat carbon steel products (slabs) and QP do not give rise to horizontally affected markets and thus those overlaps are not discussed further in this Decision.

(487)    With regard to vertical links, vertically affected markets arise (i) within the flat carbon steel value chain, (ii) between the production and supply of various flat carbon steel products and the downstream activities of the Parties, in particular the distribution of flat carbon steel and the production and supply of welded carbon steel tubes. Vertically affected markets also arise between (iii) the upstream production and supply of semi-finished products (slabs) and the downstream production and supply of HR.

9.3. Market structure and market shares of flat carbon steel products in the EEA and parts thereof

(488)    This Section reports the market shares tables for all flat carbon steel products under investigation, and provides an overview of the merged entity's combined market share. Market shares will then be discussed in more detail in each Section relating to a particular product. For the reasons explained in Section 8.2, the analysis will consider merchant sales shares, capacity shares and consumption shares. The geographic differentiation that characterises the flat carbon steel markets is taken into account by presenting market shares at different levels. Specifically, Section 9.3.1 presents the market shares at EEA level, whereas Section 9.3.2 presents the market shares for Southern Europe and parts thereof.

9.3.1. Market shares in the EEA

(489)    Table 21 presents the shares at EEA level for HR, CR, GS and its sub-segments  HDG and EG. In terms of combined sales shares in the EEA, the merged entity  would have [20-30]% in HR, [30-40]% in CR, [30-40]% in GS, [30-40]% in HDG and [30-40]% in EG.

(490)    As explained in Section 8.2, sales shares do not capture the position of integrated producers in terms of their production capabilities at each level of the value chain. The market power of the Parties and of competitors  stemming  from  their  production assets in the EEA is reflected by their capacity shares, which are reported in Table 22. The Commission provides capacity shares of the Parties in the EEA for HR, CR and GS. The figures are computed using the data provided by the main EEA competitors of the Parties on their respective capacities, complemented by the data provided by Notifying Party for smaller players. (323)

(491)    The table is divided in two panels. The figures in the top panel of  Table  22  (Scenario 1) are computed using the capacity readily available in the EEA market. That is, the capacity that would allow production to be easily ramped up readily supplying additional volumes to the merchant market or to further downstream processing. In Scenario 1, for example, the HR rolling capacity of Ilva is […] Mt,  that is the rolling capacity currently available to Ilva due to the imposition of the capacity cap on its blast furnaces.

(492)    The figures in the bottom panel of Table 22 (Scenario 2) are computed, including a capacity that is not currently available, but that could be brought to the market with limited constraints in terms of investments and business decisions. In Scenario 2, for example, the HR rolling capacity of Ilva is […] Mt, because Ilva could, in principle, purchase slabs from the merchant market and hot-roll them achieving such volume, without exceeding the environmental boundaries imposed by the Italian courts. Such a business strategy would be feasible, […]. Similar considerations apply also for the Parties' competitors.

(493)    In both scenarios, the Parties' combined capacity shares in the EEA are sizeable and exceed 40% for all products. Specifically, in HR the Parties' combined capacity share ranges between [40-50]% (Scenario 1) and [40-50]% (Scenario 2); in CR it ranges between [40-50]% (Scenario 1) and [40-50]% (Scenario 2); in GS it ranges between [40-50]% (Scenario 1) and [40-50]% (Scenario 2); in HDG it ranges between [40-50]% (Scenario 1) and [40-50]% (Scenario 2); in EG, it ranges between  [50-60]% (Scenario 1) and [40-50]% (Scenario 2).

(494)    Overall, the capacity shares show that the post-Transaction, ArcelorMittal would structurally increase and further consolidate its leadership position in the EEA  market in HR, CR, and GS.

(495)    Table 23 shows the consumption shares described in recitals (459)-(461) for HR and CR in the EEA. The table shows that a sizeable proportion of EEA HR and CR demand would be fulfilled by the merged entity post-Transaction. Specifically, [30-40]% of HR demand in the EEA would be served by the merged entity, and [40-50]% of CR demand in the EEA would be served by the merged entity.

(496)    As a robustness check, the Commission computed consumption shares for the EEA HR and CR market taking into account the approach proposed by the Notifying Party in the Reply to the SO. That is, to include in the computation the amount of imports in downstream products (CR, GS, etc.), and removing from the computation the corresponding exports of EEA suppliers. This results in slightly lower, but still sizeable, consumption shares for the merged entity in HR ([30-40]%, of which ArcelorMittal [30-40]% and Ilva [0-5]%) and CR  ([30-40]%,  of  which ArcelorMittal [30-40]% and Ilva [5-10]%).

9.3.2. Market shares in Southern Europe and parts thereof

(497)    As explained in Section 7.2.5, there are several indications that the EEA flat carbon steel   market   is   geographically   differentiated.   It   appears   that   the competitive interaction between the Parties is stronger in Southern Europe (particularly in Italy) than in Northern Europe.

(498)    This is shown in Table 24, presenting the sales shares in Southern Europe. Compared to the sales shares at EEA level, in Southern Europe the merged entity would have significantly higher sales shares – specifically, [30-40]% in HR, [40-50]% in CR, [40-50]% in GS, [40-50]% in HDG, and [50-60]% in EG.

(499)    Table 25 shows the sales shares for Italy, the largest country by consumption of flat steel products in Southern Europe and the country where the competitive interaction between the Parties is strongest. In Italy, the merged entity would have sales shares  of  [30-40]%  in  HR,  [30-40]%  in  CR,  [30-40]%  in  GS,  [30-40]%  in  HDG   and [30-40]% in EG.

(500)    The sales shares at different geographic levels in the EEA further confirm the EEA flat carbon steel markets are geographically differentiated. The merged entity would have a stronger position in Southern Europe than in Northern Europe.

(501)    Furthermore, in Southern Europe the sales shares may not fully reflect the strength of each market player in terms of market power, as explained in Section 8.2. This is illustrated by Table 26, reporting the consumption shares for HR and CR in Southern Europe, which indicates that the merged entity would serve the majority of demand  in that region. Specifically, the merged entity would fulfil [50-60]% of the HR demand and [60-70]% of CR demand in Southern Europe.

(502)    As done for the consumption shares at EEA level, the Commission made a  robustness check and computed the consumption shares in Southern Europe taking into account the approach proposed by the Notifying Party in the Reply to the SO. Thus, the Commission included in the computation the amount of imports (both from non-Southern European producers and by non-EEA producers) in downstream products (CR, GS, etc.), and removed from the computation the corresponding exports of Southern European suppliers. This results in a lower, but still sizeable, consumption shares for the merged entity in  HR  ([40-50]%,  of  which  ArcelorMittal [30-40]% and Ilva [10-20]%) and CR ([40-50]%, of which ArcelorMittal [30-40]% and Ilva [10-20]%).

(503)    In conclusion, the market shares presented in this Section, Section 9.3, indicate that the merged entity would have a sizeable market position in the EEA, and would have an even stronger market position in Southern Europe and, in particular, in Italy.

9.4. Production and supply of hot-rolled flat carbon steel products (HR): horizontal non-coordinated effects

9.4.1. Post-Transaction, the merged entity  would  be  the  largest  HR  manufacturer  in  the EEA, with increased market power

(504)    Based on the HR market shares (sales shares, capacity shares and consumption shares), the Commission finds that post-Transaction the merged entity would be the market leader, with a sizeable market position in the EEA HR market, and particularly in parts of Southern Europe (in particular Italy). This suggests that the merged entity would enjoy increased market power post-Transaction.

(505)    As discussed in Section 9.3.1, the Parties' EEA combined sales share for HR in 2016 (Table 21) would amount to [20-30]% with an increment of [5-10]%. The EEA competitors with the highest sales share would be ThyssenKrupp and Tata Steel, with a sales share of [5-10]% each in the EEA, about […] times smaller than the merged entity. In the EEA, the aggregate imports would account together for a sales share of [20-30]%. Table 24 shows that in Southern Europe, the Parties’ combined sales share would be [30-40]% with an increment represented by Ilva's share of [10-20]%. The second largest EEA supplier in Southern Europe in terms of merchant sales would be Arvedi, with sales shares of [10-20]%, more than […] times lower than the merged entity. At the same time, the aggregate imports would account for a sales share of [30-40]%. As regards Italy, the largest flat carbon steel market in Southern Europe and the area where the competitive interaction between the Parties is the most prominent, the combined market share of the Parties would amount to [30-40]% with an increment of [10-20]% resulting from ArcelorMittal's share (Table 25). The second largest EEA supplier would be Arvedi in Italy with a sales share of about […] of the merged entity ([10-20]%), and Tata Steel, with a sales share about […] times smaller than the merged entity ([40-50]%). The aggregate imports, however, would have a sizeable sales share of [30-40]% in Italy.

(506)    Table 27 and Table 28 show the sales shares from 2011 to 2016 in the EEA and in Southern Europe, respectively. The imposition of the  capacity cap on  Ilva  from  late 2012 resulted in a decrease  in  its  sales  shares  from 2013  onwards,  both  in the EEA and in Southern Europe. However, the tables show that in 2011 and 2012 Ilva was the second-largest HR supplier in the EEA after ArcelorMittal, and the largest HR supplier in 2012 in Southern Europe. The tables also show the steep increase in HR sales shares of imports in 2015–2016, a period characterised by dumped HR  volumes  from  non-EEA  producers,  as  discussed  in  details  in Section 9.5.5. Lastly, Table 28 shows that EEA suppliers not based in Southern Europe had a limited presence in Southern Europe throughout the period 2011–2016. Most notably, Tata Steel and ThyssenKrupp, the two largest HR suppliers in the EEA after ArcelorMittal in recent years, had each a sales share lower than [0-5]% throughout 2011–2016 in Southern Europe. Moreover, the combined sales share of suppliers  not  based  in  Southern  Europe  was  never  above  [10-20]%  in  Southern Europe.

(507)    As shown in Table 22, the Parties' combined capacity shares at EEA level in 2016 in HR would range between [40-50]% (Scenario 1) and [40-50]% (Scenario 2), with an increment represented by Ilva of [5-10]% for Scenario 1 and Scenario  2,  respectively.

(508)    The Parties' EEA combined HR consumption share would amount to [30-40]% with an increment of [5-10]% (Table 23), whereas the Parties' combined consumption share  in  Southern  Europe  would  amount  to  in  [50-60]%  with  an  increment of [20-30]% (Table 26).

(509)    The market shares further show a significant asymmetry in market positions between the merged entity and other EEA competitors, which reflect sizeable differences in their production capabilities. In the EEA, the EEA supplier with the highest consumption share after the Parties would be ThyssenKrupp, with a consumption share of [10-20]%. Thus, the merged entity would have a consumption share […] higher than the second largest EEA competitor. In Southern Europe, the EEA competitor with the highest consumption share would be Arvedi, with a consumption share of [10-20]%. Thus, the merged entity would have a consumption share […] higher than the second largest EEA competitor. Moreover, in Southern Europe the consumption share of the aggregate imports would be [20-30]%, less than […] of the merged entity. Thus, the merged entity would account for the highest consumption share both in the EEA and in Southern Europe.

(510)    In a nutshell, post-Transaction the merged entity would be the largest HR manufacturer in the EEA, with the HR production capabilities of the other EEA producers being significantly smaller. The merged entity, in particular, would be approximately […] as large as the second and third largest EEA competitors, Tata Steel and ThyssenKrupp, at the EEA level, both when considering sales shares and consumption shares. With the exception of imports, which will be examined in more detail in Section 9.4.5, the remaining players would account for a share that would be several times smaller than that of the merged entity.

(511)    Moreover, post-Transaction the merged entity would be in an even stronger position in Southern Europe, where it will be the only integrated EEA supplier with significant capacity and production capabilities. Other EEA competitors based in Southern Europe are Arvedi, which is an integrated supplier but has limited capacity and a different technology that may  hinder  the  quality  of  its  products  (see  Section 5.3.1.3);  Metinvest,  which  does  not  own  primary  steel  production  in  the EEA (328) and has only a limited production capacity of less than […] Mt; and Marcegaglia, the largest EEA re-roller, which however does not own primary steel production (crude steel) and hot-rolling lines, and hence is dependent on other steel manufacturers (including the Parties) and/or imports to feed its CR and HDG/EG production.

9.4.2. ArcelorMittal and Ilva are close competitors in the production and supply of HR

9.4.2.1. The Notifying Party's views

(512)    ArcelorMittal submitted that the Parties are not close rivals and their flat carbon steel businesses are largely complementary. The Notifying Party argued that the steel assets of ArcelorMittal and Ilva are geographically distant and complementary. According to ArcelorMittal, its business focuses on supplying specialty products to automotive and other demanding customers, whilst Ilva's business is focused on the production of commodity grades. As a result of these alleged differences, ArcelorMittal submitted that (i) there is a limited overlap in the customers of ArcelorMittal and Ilva and (ii) such overlap is largely limited to the production and supply of commodity flat carbon steel products, predominantly in Southern Europe.

(513)    The Notifying Party further submitted in its Reply to the SO that other EEA and non-EEA suppliers can, and do, produce all of the Parties' overlapping commodity products, while many of them regularly supply into Southern Europe.

(514)    Furthermore, in its Reply to the SO, the Notifying Party contested the Commission's quantitative switching analysis which found that ArcelorMittal gained significantly from Ilva's reduced output, caused by the imposition of its capacity  cap. ArcelorMittal submitted its own supply shock analysis and an analysis of switching behaviour at the customer level, from which it concluded that imports, rather than ArcelorMittal, have gained most from Ilva's reduced output.

(515)    Finally, in its Reply to the SO, the Notifying Party submitted that the Parties' geographic footprint is not indicative of close competition. Rather, it is argued that the HR assets of Northern European suppliers are situated as close or closer to the centre of Southern European demand as Ilva's Taranto and that the majority of Southern European demand is located in either the north of Italy or the north of Spain.

9.4.2.2. The Commission's assessment

(516)    Based on the results of its market investigation, the Commission considers that ArcelorMittal and Ilva are close competitors in the production and supply of HR. The finding is based in particular on the fact that both companies are, together with imports, the main suppliers of hot rolled steel, particularly in commodity grades, in the South of Europe. The finding of closeness is based on the following elements:

(a)   Ilva's product offering directly competes with ArcelorMittal's and imports' product offering;

(b)   ArcelorMittal's and Ilva's geographic footprints in the EEA significantly overlap and both are integrated steel producers using a BOF technology;

(c)   ArcelorMittal's, Ilva's and imports' customers overlap in the areas where the Parties' compete against each other;

(d)   [Internal document];

(e)   The    Commission's    quantitative    analysis    finds    significant          gains                  for ArcelorMittal following the imposition of the capacity cap on Ilva;

(f)   Feedback from market participants confirms that ArcelorMittal and Ilva compete closely with each other, as well as with imports.

(517)    It should be noted that, while this Section focuses on HR, a large part of the evidence presented is applicable also to other downstream products. For example, the evidence from ArcelorMittal's internal documents generally refers to [internal document].

(518)    The finding that the Parties are close competitors does not preclude the fact that other players exert a competitive constraint, in particular as regards the production and supply of commodity HR products in Southern Europe and in Italy. In terms of closeness of competition, therefore, aggregate imports can be considered to be close competitors of the Parties. However, imports face a number of limitations that undermine their ability and incentives to defeat the likely price increase by the merged  entity  post-Transaction.  Such  limitations  are  further  discussed  in  Section 9.4.5.

9.4.2.3.  Ilva's product offering directly competes with ArcelorMittal's and imports' product offering

(519)    Contrary to the alleged complementarity of ArcelorMittal's and Ilva's HR product offerings argued by the Notifying Party, the Commission considers that the Parties' product offerings largely overlap. This is for the following reasons.

(520)    First, for technical and commercial reasons, steel manufacturers cannot only produce specialty steel products, but they also have to produce commodity products.

(521)    As  described  in  recital  (294),  for  technical  reasons,  it  is  not  possible  to dedicate 100% of the production capacity to specialty products, which require specific process conditions. Periods of 'relaxation' are required at all stages, from casting, through rolling, to finishing. For example, not all slabs derived from the  same caster sequence can be used to produce specialty products and worn cylinders cannot be used to roll specialty products. Due to these so-called 'production mix considerations/limitations', an integrated steel manufacturer needs to produce certain amounts  of commodity grades  together with  high-end  products. (329)  The   Notifying Party reported that for ArcelorMittal‘[s]labs for automotive exposed parts account for at most […]% of the output of a continuous caster.’ or that ‘[m]ost thin gauges account for at most […]% of the output of a hot-rolling line’. (330) Such technical constraints were also confirmed in the Commission's market investigation by the Parties' competitors. According to one competitor: '[…] from a technical perspective, steelworks also cannot de facto only produce high quality steel as the equipment wears quickly when producing high-quality steel and would require to be replaced/maintained more often. For instance, the production of many  high-end steels require a perfect rolling cylinder (with a perfect plain surface) to create a high-quality surface for the steel product. Once the cylinder is worn, it can no longer be used for the production of high-end products that require good surface quality  and small tolerances; however, it can still be used to produce more commodity products. While it could technically be possible to always replace the cylinder once it is slightly worn and continue producing high-end products, that would be very expensive. Thus, in practice, ‘relaxing’ production campaigns need to be run and some commodity grades must de facto be produced during the lifetime of each equipment installation to keep the operations economically sound.' (331) Another competitor stated that ‘[i]n practice the production of some commodity grades is required. […] the production of value-added products is more intense and demanding on the production equipment and, therefore, 'relaxation' periods are needed’. (332)

(522)    Furthermore, due to commercial reasons, producers may shift production between lower and higher grade products depending on market demand. The Parties' competitors also support this view. One competitor explained: ‘Steel makers often fill their mills with lower grade products if the demand for high end steel is not enough.' (333) Consequently, a competitor considered that also ArcelorMittal was active in the production and supply of commodity grades ‘in order to utilize their production capacity’. (334) Similarly, another competitor had doubts that a business model based on the production of, what it refers to as, 'differentiated' products could remain profitable and competitive. (335)

(523)    Second, commodity products account for a significant proportion of both Ilva's and ArcelorMittal's HR sales. The […] of Ilva's HR sales in the EEA are made of commodity products (as shown in Figure 56), and about […] of ArcelorMittal's HR sales in the EEA are made of commodity products. (336) Figure 57 illustrates the split of ArcelorMittal's HR sales between specialty products and commodity products in the EEA. For the time period 2014–2016, […] of ArcelorMittal's HR sales consistently consisted of commodity products, while […] of its sales consisted of specialty products. Figure 58 reproduces the same split for Italy only, and shows that commodity grades still represent a […] proportion of about […]% of ArcelorMittal's sales in Italy. The feedback from the Commission's market investigation also points to the product offering of imports being for the vast majority of commodity products, similar to Ilva's.

Figure 56 – Ilva total flat steel shipments of HR in the EEA

[…]

Note: 2017 only January-June

Source: Commission's computation based on data provided by ArcelorMittal (Response to RFI 5)

Figure 57 – ArcelorMittal total flat steel shipments of HR in the EEA

[…]

Note: 2017 only January-June

Source: Commission's computation based on data provided by ArcelorMittal (Response to RFI 5)

Figure 58 – ArcelorMittal total flat steel shipments of HR in Italy

[…]

Note: 2017 only January-June

Source: Commission's computation based on data provided by ArcelorMittal (Response to RFI 5)

(524)    Third, while ArcelorMittal's product offering in terms of steel grades is much broader than Ilva's, the Commission notes that all of the […] EN grades of HR sold by Ilva are also sold by ArcelorMittal, and that the sales for those […] grades account for about […]% of ArcelorMittal's sales in the EEA (the sales of these […] products amounted to […] Kt in 2016 whereas the total sales of specialty products amounted  to […] Kt). (337) The Parties' overlap in terms of sales of common grades is therefore sizeable. In particular, Table 29 reports the Parties' sales of the steel grades that account for more than […]% of the total sales. The table shows that […]. For example, the sales of […] and […] account for […]% of Ilva's HR sales and […]% of ArcelorMittal's HR sales in the EEA, […] kt and […] kt respectively. Precise figures on the grades offered by imports are not available, however the Commission considers it likely that non-EEA producers exporting to the EEA are able to produce and export the most commonly traded HR grades, such as […] and […].

(525)    Fourth, as indicated in Figure 59, ArcelorMittal itself observed that, out of all the products that it supplies in Italy, only a small portion ([…] of the […] of products delivered, that is […]%) cannot be produced by Ilva, and such a […] refers to all flat carbon steel products, including those that Ilva does not manufacture ([…]).

Figure 59 – ArcelorMittal internal document (338)

[…]

(526)    Fifth, in the Commission's view the competitive interactions between ArcelorMittal and Ilva are not limited to commodity products, despite Ilva's production being focused predominantly on commodity products. This is because, as explained in Section 6.10, the Commission considers that commodity and specialty products belong to the same relevant product market, and that competition in commodity products will also affect competition in specialty products due to, for example, the pricing scheme adopted in the flat carbon steel industry (that is ‘base price + extras’), which results in a competitive interaction between commodity and specialty  products.

(527)    Lastly, the Commission does not contest, as the Notifying Party appears to claim in its Reply to the SO, that other suppliers (domestic and non-EEA) also can and do produce commodity grades. In particular, the market investigation confirmed that non-EEA suppliers predominantly export commodity HR products to the EEA,  which are in direct competition with Ilva's and ArcelorMittal's commodity products. However, the actual and potential overlap between the product offerings of  the Parties supports the finding of closeness, in contrast with the claims of the Notifying Party about complementarity in the product offerings of ArcelorMittal and Ilva.

9.4.2.4. ArcelorMittal's and Ilva's HR geographic footprints largely overlap

(528)    The Commission considers that ArcelorMittal and Ilva compete closely also in view of the location of their manufacturing facilities. (339) They largely overlap in southern Europe, where the geographic distribution of imports within the EEA is also more focused.

(529)    First, as shown in Figure 32 ArcelorMittal and Ilva operate the largest HR production facilities in Southern Europe. Furthermore, ArcelorMittal's plant in  Fos-sur-Mer ([…] Mt of HR capacity) is one of the closest HR production facilities to Ilva's plant in Taranto, and distributes HR also for further processing to ArcelorMittal's other plants in Italy. In this respect, a customer active in distribution primarily in Italy also stated that the steel purchased from ArcelorMittal, mainly black or pickled HR products, 'comes primarily from ArcelorMittal's plant in Fos-sur-Mer in southern France.' (340) As regards ArcelorMittal's comparison of the distance between Ilva and other European producers' HR assets to what they consider the centres of demand in Southern Europe, it is important to point out that this comparison is based on driving distance. As Ilva can supply from its Taranto plant to the North of Italy by sea, this calls into question the validity of such comparison. The Commission's finding is further evidenced by the limited market shares of other European producers in the South of Europe, as observed in Table 24 and Table 26. Furthermore, the Notifying Party's comparison based on driving distance fails to recognise that its own plant in Fos-sur-Mer remains the closest to this centre of demand.

(530)    Second, most of the other EEA competitors with sizeable HR capacities have production facilities located in Northern Europe. For example, Salzgitter and ThyssenKrupp's main production assets are located in Germany, while Tata Steel is located in the Netherlands and in the UK. Contrary to ArcelorMittal's claim that Northern European suppliers can and do supply into Southern Europe, it appears that the presence of such Northern European suppliers in Southern Europe is limited. Feedback  from  Marcegaglia,  the  largest  and  most  prominent  re-roller  in Italy, indicates that sourcing from these Northern European suppliers is generally constrained by logistic costs (especially by land) and a lack of HR offered to Italy. (341) The presence of Northern European suppliers in Southern Europe is characterised by limited sales shares, as observed in Table 24 and Table 26. Furthermore, it is evident from Figures 36–38 that only a limited part of the HR production of Northern European suppliers is sold in Italy. As shown in Section 9.3 presenting sales shares, the only sizeable presence in Southern Europe other than the Parties are imports.

(531)    Third, other EEA competitors with HR capacities located geographically close to  Ilva do not appear to exert the same degree of competitive pressure as the  competitive constraint that the Parties exert on each other. In particular, Arvedi, which is located in Italy, despite ArcelorMittal claiming that it is able to compete aggressively with the Parties, does not appear to be regarded by market participants  as equally close to Ilva as ArcelorMittal. The main reasons indicated by customers and competitors are that Arvedi has a more limited HR product range and the company uses a different steel production technology, namely the Electric Arc Furnace. One of the Parties' competitors noted that 'Ilva is a more important supplier than Arvedi because Ilva has a bigger range of products. Arvedi can only offer two different widths, whereas Ilva can offer any width of products.' (342) A customer active in distribution also indicated that 'Arvedi's portfolio is not as wide as ArcelorMittal's or Ilva's. In particular, Arvedi cannot supply all the physical sizes of coils that ArcelorMittal or Ilva supplies. Arvedi only supplies 1.25 and 1.5m wide coils and no other widths, which is a significant limitation. Quality-wise Arvedi does also not in all applications perform as good as other suppliers such as ArcelorMittal or Ilva.  The reason for this is that Arvedi employs electric arc furnace process instead of an integrated blast furnace – basic oxygen furnace route […].' (343)

(532)    As regards Marcegaglia, the other large steelmaker with facilities in Italy, it is not regarded as a close competitor in view of the fact that it is a re-roller which depends on external sourcing of HR coils to manufacture downstream products.

(533)    Fourth, the geographic closeness between the Parties is also reflected in ArcelorMittal's HR sales on the Italian market. In the period between 2011 and 2016 ArcelorMittal's HR sales in Italy accounted for […]% of the total sales made by EEA suppliers not located in Italy. (344) Contrary to the Notifying Party's arguments in the Reply to the SO, this shows that, among EEA suppliers, ArcelorMittal exerts a significant competitive constraint in the country despite not having a production facility in Italy. This finding is not invalidated by the claim that important import volumes land in Italy.

(534)    Fifth, feedback from market participants indicates that the geographic footprint of ArcelorMittal's and Ilva's HR sales overlaps in Italy and in its surroundings. (345) A customer for instance pointed out that ‘ILVA and AM are the only European mills supplying regularly the south european market, with a big range of products (HR, HDG, EG, CRC, HRP)’. (346) Another customer indicated that ‘both mills compete for supplies to our Italian factories on commodity grade steel’. (347)

9.4.2.5. ArcelorMittal's, Ilva's and imports' customers overlap in the areas where the Parties compete against each other

(535)    On the basis of the market investigation, the Commission observes that closeness of competition between the Parties also follows from an analysis of the customer overlaps between the Parties. Based on the available evidence, the Commission considers in particular that the areas where Ilva is present, that is Southern Europe and Italy, are characterised by a significant customer overlap and customer switching between the Parties, as well as with imports, which indicates an intense competitive interaction between ArcelorMittal, Ilva and imports at least in some significant parts of the EEA.

(536)    The Notifying Party submitted that only […]% of ArcelorMittal's HR sales  are related to customer locations common to Ilva's sales ([…]% considering customer groups). (348) On the other hand, considering the customer base of Ilva, […]% of Ilva's HR sales are related to customer location common to ArcelorMittal ([…]% considering customer groups).

(537)    Further submissions by the Parties still show a sizeable customer overlap between  the Parties and customer switching from Ilva to ArcelorMittal and to imports, in particular in Southern Europe and in Italy. (349) Moreover, the Commission considers that ArcelorMittal's customer overlap analysis might underestimate the competitive pressure exerted by the Parties on each other.

(538)    In the first place, even if one of the Parties does not conclude any sale to a specific customer, it might still exert a competitive constraint on the other Party if it is considered a potential supplier in the same area. This effect is not captured by the Notifying Party's analysis. It is, however, relevant as the geographic footprints of the Parties largely overlap in Southern Europe and they are thus potential suppliers in the same area (see Section 9.4.2.4).

(539)    In the second place, the Commission considers that in light of ArcelorMittal being a considerably larger company than Ilva, it is not appropriate to base a customer overlap analysis on ArcelorMittal's sales. Only the customer overlap analysis based on Ilva's footprint can provide meaningful insights for the assessment of closeness of competition between the Parties. On that basis, the analysis carried out by the Notifying Party indicates that a large proportion of Ilva's HR volumes are sold to common customers ([…]% including Marcegaglia and […]% excluding Marcegaglia), and that there is a sizeable amount of customers that switch from Ilva to ArcelorMittal.

(540)    Taking into account the limits of this analysis in capturing the competitive constraints, the Commission considers that the overlap in terms of customers indicates that the Parties compete closely against each other and against imports in the areas where their geographic footprints overlap.

9.4.2.6. ArcelorMittal's internal documents indicate that the Parties closely compete

(541)    The Commission considers that ArcelorMittal's internal documents describe a consistent pattern of volumes lost and gained to and from Ilva which indicates a high degree of closeness of competition between the Parties.

(542)    First, [internal document]. Furthermore, it should be noted that among the competitors analysed by ArcelorMittal, only those with installed capacity in the EEA are present, whereas imports are not considered.

Figure 60 – ArcelorMittal's internal document dated December 2014 (ID1660-028435)

[…]

(543)    Second, [internal document].

(544)    The Notifying Party's argument in its Reply to the SO (350) that the actual sales share loss of Ilva in the period 2013–2014 would actually decrease by […] p.p. is factually wrong. From the data available to the Commission, Ilva's sales shares loss in Italy in the period 2013–2014 for all flat products amounts to […] p.p., as indicated in ArcelorMittal’s internal document. Furthermore, the Notifying Party claimed that the Commission misread the table by attributing the volume loss from Ilva and Arvedi to ArcelorMittal and imports. As a matter of fact, the Commission simply quotes ArcelorMittal's internal document.

Figure 61 – ArcelorMittal internal document dated April 2015 (ID1659-073497)

[…]

(545)    The same document shows [internal document].

Figure 62 – ArcelorMittal internal document dated April 2015 (ID1659-073497)

[…]

(546)    Third, an internal document [internal document].

Figure 63 – ArcelorMittal internal document dated October 2015 (Attached to ID1660-41054) (351)

[…]

(547)    This is further supported by the boxes on the right-hand side of Figure 63. […].

(548)    The same document presents more disaggregated figures in a slide reported in Figure 64, which explains the simulation of ArcelorMittal in more detail. […].

Figure 64 – ArcelorMittal internal document dated October 2015 (Attached to ID1660-41054)

[…]

(549)    Another ArcelorMittal document, as reported in Figure 65 assesses the alternative scenarios for the re-activation of Ilva's capacity and quantifies their respective impact on ArcelorMittal's deliveries in 2015–2016. The analysis concludes that a  reactivation of BF5 would result [internal document].

(550)    The Commission considers that the Notifying Party's argument in its Reply to the SO whereby the market developed in a different manner and Ilva's BF5 was not actually restarted, is immaterial for the assessment of ArcelorMittal's internal documents in relation to the closeness of competition between the Parties.

Figure 65 – ArcelorMittal internal document (352)

[…]

(551)    Fourth, a document [internal document]. (353)

Figure 66 – […] (354)

[…]

9.4.2.7. The Commission's quantitative analysis finds significant gains for ArcelorMittal following the imposition of the capacity cap on Ilva

(552)    The Commission carried out a quantitative analysis to investigate the closeness of competition between the Parties, and to gather insights in relation to the competitive constraint exercised by imports on the EEA HR flat steel market. A complete description of the Commission's quantitative analysis is provided in the Annex 1 to this Decision. (355)

(553)    In relation to the closeness of competition between the Parties, the Commission's quantitative analysis supports the findings gathered during the Commission's market investigation that ArcelorMittal and Ilva are close competitors.

(554)    The analysis estimates the losses and gains in HR sales shares by market participants caused by the imposition of the capacity cap on Ilva from late 2012. The analysis finds that in Southern Europe (356) and in Italy, [commercial information].

(555)    The magnitude of the effect estimated in the Commission's analysis is in line with and corroborated by ArcelorMittal's internal calculations of volumes losses and gains to and from Ilva described in Section 9.4.2.6.

(556)    Inferences in relation to the period 2015–2016 are difficult to make as two events happened simultaneously: (i) a further reduction of Ilva's capacity due to the mothballing of BF5, and (ii) dumping practices from non-EEA suppliers exporting to the EEA. The analysis estimates a significant increase in the sales shares of imports and a decrease in the sales shares of Ilva during this period. This would point to a significant diversion from Ilva to imports. The two simultaneous events, however, have opposite effects on the sales shares of some market participants, most notably   of ArcelorMittal. As the analysis is not able to disentangle the effects of the two events on the market participants' sales shares, which are hence estimated jointly, inferences regarding closeness of competition are difficult to draw from the estimates in this period. However, in light of the qualitative evidence described in this Section which point to an intense competition between Ilva, ArcelorMittal and imports in Southern Europe, the Commission cannot exclude that during the dumping period, diversion of Ilva sales in Southern Europe and Italy also benefitted imports.

(557)    Overall, the quantitative analysis shows that, at least for the period in which no market disturbances were present (that is prior to the dumping period), the imposition of the capacity cap on Ilva determined a sizeable loss in Ilva's HR sales shares in Southern Europe and in Italy, [commercial information]. This finding indicates that the Parties are close competitors and supports the qualitative evidence mentioned above in Sections 9.4.2.2 to 9.4.2.6.

(558)    The Notifying Party's criticisms to the Commission's quantitative analysis included  in its Reply to the SO do not change the Commission's conclusion, and are addressed in the Annex 1 to this Decision. (357)

(559)    The Notifying Party further submitted an analysis of switching behaviour from which it concluded that imports were Ilva's closest competitor. The Commission does not put into question the competitive constraint exerted by imports on aggregate and  their closeness of competition with the Parties in Southern Europe and in Italy. This, however, does not undermine a finding that the Parties are close competitors to each other.

(560)    Moreover, the Commission notes the following. First, as explained in the Annex 1 to this Decision, it is important to take into account factors determining the availability of imports in the EEA. Second, the Notifying Party's analysis focussed on the period between 2015 and 2016, which was characterised by dumped volumes from  non-EEA exporters. Third, when cross-checking with sourcing information obtained directly from customers, the Commission found that the Notifying Party's analysis overstated the extent to which customers resorted to imports. (358)

(561)    In addition, the Commission notes that the recent trade measures imposed on the products concerned from specific non-EEA countries, as described in Section 5.3.3.3 and Section 9.4.5.7, appear to have removed the market perturbation and thus reduced the undue pressure exerted by imports in the recent years of market  distortion resulting from dumping prices.

(562)    The Commission further considers that imports will not be able to sufficiently constrain the merged entity, as explained in Section 9.4.5.

9.4.2.8. Feedback from market participants confirms that Ilva and ArcelorMittal are close competitors

(563)    The market feedback received during the Commission's market investigation confirms that the Parties are close competitors.

(564)    First, a large number of customers suggested that ArcelorMittal and Ilva are close competitors in terms of product offering. When asked about the competitive interaction between the Parties in Italy and its surroundings, a very large majority of steel service centres and other distributors considered the Parties to be close competitors in the production and supply of HR commodity grade products. (359) A majority of those customers also considered the Parties as close competitors in the production and supply of HR products of non-standard dimensions. (360) Similarly, industrial customers that responded to the Commission's investigation also saw the Parties as close competitors, particularly in commodity grade HR products in Italy and its surroundings. (361) For instance, an industrial customer stated that 'Arcelormittal and Ilva are both active on part of the same market of commodities products and compete closely with each other in relation to HR products.' (362)

(565)    Second, a large majority of industrial customers indicated that ArcelorMittal and Ilva were close competitors also on the price and range of HR products. (363) One customer active in the distribution market stated that 'ILVA and AM are the only European mills supplying regularly the south european market, with a big range of products (HR, HDG, EG, CRC, HRP)' (364) while another customer confirmed that 'AM has a wider product portfolio and bigger possibilities for differentiation in non- commodities. In commodities, on the contrary, AM’s and Ilva’s products are comparable.' (365) Furthermore, a very large number of the Parties' competitors also indicated ArcelorMittal and Ilva as close competitors on price and range of products. One competitor explained: 'In Southern Europe AM and Ilva compete strongly in commodity hot rolled products.' (366)

9.4.2.9.  Conclusion as regards closeness of competition between ArcelorMittal and Ilva

(566)    Based on the evidence described in this Section, the Commission considers that the Parties are close competitors, in particular in Southern Europe and in Italy where  their activities mostly overlap. The evidence also indicates that imports have a sizeable presence in Southern Europe and compete closely with the Parties in the commodity products segment where Ilva, in particular, is active.

9.4.3.  Ilva is an important competitive force

9.4.3.1. The Notifying Party's views

(567)    ArcelorMittal claimed that there is no basis to consider Ilva an  ‘important competitive force’ within the meaning of the Horizontal Merger Guidelines, which typically associate this concept with aggressive new entrants or innovators. ArcelorMittal relied on the following arguments.

(1)             Ilva is capacity-constrained and has reduced capacity by […] Mt. Therefore, it cannot significantly increase output to constrain rivals.

(2)             Ilva's market share has significantly decreased. According to ArcelorMittal, the finding that a firm represents an ‘important competitive force’ is almost invariably associated with increasing market shares.

(3)             [Parties' submissions].

(4)             [Parties' submissions].

(5)             [Parties' submissions].

(6)             The development of the spread between the Northern and Southern European HR price is not driven by Ilva.

(7)             ArcelorMittal predominantly loses volumes to other suppliers, notably imports.

(568)    In its Reply to the SO, the Notifying Party argued that the SO overstated the Transaction's competitive impact by asserting that Ilva was an ‘important  competitive force’. The Notifying Party reiterated that this was not the case, given that Ilva has commercially and financially collapsed. More particularly:

(1)             [Parties' submissions].

(2)             [Parties' submissions].

9.4.3.2. The Commission's assessment

(569)    For the reasons set out in this Section, the Commission considers that Ilva is an important competitive force in the EEA HR market, and particularly in Southern Europe and Italy.

(570)    The Commission observes that the prospect of a market share increase is not necessary to conclude that Ilva is an important competitive force. The factors, described in the following Sections 9.4.3.3 to 9.4.3.6, may be sufficient to lead to the conclusion that Ilva has a higher impact on the competitive process than its current market share, or similar measures, would indicate.

(571)    Further, the body of evidence described in this Section, as well as the other evidence described in this Decision, in particular those in Sections 9.4.2 and 9.4.6, indicate  that Ilva represents an important competitive constraint in the EEA HR market. Such important competitive constraint would be eliminated by the proposed Transaction.

9.4.3.3. Ilva has significant capacity which can impact EEA markets

(572)    First, Ilva is the largest steel producer in Italy, and the only one in Italy that has an integrated primary steel production that uses basic oxygen furnace technology. Its plant in Taranto is the largest single-site plant in the EEA, with installed  slab capacity of […] Mt, and an HR capacity of […] Mt. Together with its satellite plants in Genova and Novi Ligure, Ilva has a combined CR capacity of […] Mt, HDG capacity of […] Mt, EG capacity of […] Mt (for a combined GS capacity of […]  Mt),  TP capacity  of  […]  Mt,  and  QP  capacity  of  […]  Mt. (367)    By  means  of comparison, Ilva's installed HR capacity of […] Mt accounts for around […]% of total EEA HR capacity. (368)

(573)    Second, even though its primary steel (slab) production is limited due to compliance with environmental regulation, Ilva remains amongst the largest EEA producers. At current production levels, Ilva accounts for […]% of EEA readily available HR capacity in 2016 and […]% (369) in terms of HR consumption, and thus remains the fourth largest steel producer in the EEA, ahead of VoestAlpine and US Steel  Košice. ArcelorMittal commented that […]. (370)

(574)    Even if its current court-imposed caps constrain Ilva's primary steel/slab capacity, Ilva's capacity and market shares would likely be expanded under any alternative buyer. (371)

(575)    Third, Ilva's ability to impact the EEA markets for flat carbon steel was best demonstrated when, due to environmental compliance issues, it had to significantly scale down the production of flat steel. [Internal document], (372) […].

(576)    In the first place, as demonstrated in Figure 67, the developments concerning the utilisation of Ilva' blast furnaces (BF5/BF1) were the single most important factors impacting the overall EEA capacity in the period from 2014–2016.

Figure 67 – Capacity changes in the EEA (373)

[…]

(577)    In the second place, the reduction of Ilva's output seems to have had a discernible impact on price levels, especially in Southern Europe.

(578)    [Internal document]:

[…] (374)

[…] (375)

[…] (376) […]. (377)

[…]. (378) […] (379)

[…] (380)

[…] (381)

(579)    In the third place, the statements presented in recital (578) relate to the withdrawal of […] Mt, and not the entire Ilva capacity from the market. Such […] Mt capacity was in itself considered as a significant factor affecting commercial dynamics. This suggests that the current capacity of Ilva of […] Mt (in HR), which is more than […] times the size of the withdrawn capacity, represents an even more significant constraint in the market.

(580)    In the fourth place, the above developments related to Ilva's capacity reductions (382), are also reflected in the evolution of the difference between HR coils prices in Northern and Southern Europe. The Commission observes that in the period between the end of 2012 and until mid-2015, […], the difference between the Northern and Southern price indices for HR coil was more stable and at lower levels as compared to the period before and after.

Figure 68 – Difference between Northern and Southern HR price indices

[…]

Source: SBB data submitted by the Notifying Party

(581)    [Commercial information].

(582)    [Internal document], (383) […]. (384)

Figure 69 – Lead times per product ArcelorMittal

[…]

(583)    [Commercial information, internal document].

(584)    In the second place, while certain customers claimed that the service levels of Ilva have deteriorated, other customers explained, for example, that […] (385), 'supplier (Ilva) which was reliable and competitive'. (386)  [Internal document]. (387) […].

9.4.3.4. Ilva has historically followed an aggressive pricing policy

(585)    First, Ilva has applied a pricing strategy based on giving priority to higher volumes of HR as well as other flat carbon steel types rather than trying to increase its prices on the market. [Internal document]. (388) [Commercial information]. (389) Beyond ArcelorMittal, Ilva also constrains other rivals especially for those products and geographic markets in which Ilva's presence is the highest.

(586)    Second, it appears that, in the past, Ilva was considered as [internal document]. (390)

(587)    Third, Ilva’s aggressive price policies have been observed by a number of flat carbon steel producers. A flat carbon steel producer explained, for instance that 'Ilva price offers,  especially  if  on  the  downside  of  the  market,  can  define  customers’ price expectation for other suppliers’. Another one reported ‘Ilva can often be referenced as the market price even when their prices are lower than seen elsewhere in Europe’. A third  one observed  that  ‘suppliers  who  usually offer  lower  prices  include Ilva,

Dunaferr and some Eastern European suppliers.' (391)

(588)    Fourth, a majority of SSCs expressing a view on the question considered that Ilva follows an aggressive price strategy. (392) Customers, for example, commented that:

'in the past, [Ilva had] aggressive price policy in the entire territory';

'Ilva could use a price agressive strategy in all products depending from the need to sell or cover plants' capacity. But it is not always like this.';

'we think that Ilva has an aggressive price policy for the above mentioned products when it sells in Italy' (393);

(589)    Fifth, similarly, roughly half of the industrial customers expressing a view on the question considered that Ilva was aggressive on prices. While a number of customers commented that they did not consider Ilva as ‘aggressive’, there appears to be an even broader understanding that Ilva's prices were lower than those of other  European suppliers. (394) For instance, a customer stated that 'based on our experience, Ilva offers prices in the lower range of the market but cannot be defined as aggressive'. Another customer submitted that 'Ilva is cheaper than other European steel suppliers but we do not consider their strategy "aggressive"; it is definitely cheaper, which is vital for us to remain competitive in the industry against European competitors (they are located in central Europe, so close to the main European mills) but the Ilva pricing policy is not that cheaper to be considered at a "dumping price" level'. A third one concluded that ‘Ilva is the most competitive commodity  supplier.' (395)

(590)    Sixth, the aggressive and volume-focussed strategy followed by Ilva is not explained exclusively by the circumstances of the financial and environmental crisis currently affecting Ilva.

(591)    In the first place, the volume focus of Ilva is a historic feature of its commercial strategy (that is it predates Ilva’s financial and environmental problems). As confirmed by certain customers, the aggressive pricing of Ilva does not only coincide with Ilva's regulatory issues and the ensuing financial difficulties. On the contrary, they explain that a low price policy was engrained in the company's business model. An SSC thus explained that Ilva has historically had a moderating role in the Italian market: 'from the mid nineties on, Ilva had a moderating function on the Italian market to the benefit of final users and their exports.' (396)  Another SSC associated price aggressiveness only with Ilva's past: 'in the past, [ilva had] aggressive price policy in the entire territory'.

(592)    In the second place, the volume-focussed strategy followed by Ilva is not inherently unprofitable as Ilva was able to generate profits before the imposition of a cap. (397) In this respect, [Parties' submissions]. (398) […], the Commission considers that historically, Ilva was generating revenue capable of sustaining CAPEX into environmental and industrial projects to sustain the scale and scope of its operations.

(593)    In the third place, as explained in Section 5.2.2.2 (399) the imposition of the cap has  been the principal source of financial losses of Ilva. Given its production capacities, Ilva has to bear the fixed costs for the production of larger quantities than it is currently able to manufacture. In this respect, [Parties's submissions] (400) […], (401) […]. (402)

(594)    Seventh, at any rate, [internal document]. (403) […]. […]. (404)

[…]. (405)

[…]. (406)

[…]. (407)

Figure 70 – […] (408)

[…]

(595)    Moreover, [internal document]. (409)

(596)    Eighth, [internal document]. (410)

(597)    Nevertheless, the data in Figure 71 indicates that […].

Figure 71 – ArcelorMittal and Ilva price premium over imports for HR

[…]

Source: Commission's calculations on the basis of Parties data (Reply to RFI 26 for Parties' prices and RFI 23 Annex 6.1 for imports prices)

9.4.3.5.  Ilva is capable of impacting rivals' commercial policy

(598)    As illustrated above, respectively in Sections 9.4.3.3 and 9.4.3.4, Ilva combines very significant capacity and a sales strategy focused on volume/turnover, rather than price/margin. For this reason, Ilva is able to constrain its rivals more than an analysis merely based on its market shares would suggest, in particular in the  Southern Europe and Italy.

(599)    The following internal document in figure 72 shows that [internal document].

Figure 72 – ArcelorMittal internal document (411)

[…]

(600)    Finally, Figure 73 shows that […].'

Figure 73 – ArcelorMittal internal document (412)

[…]

(601)    Similarly, a customer explained that Ilva was capable of impacting the commercial strategy of the other three largest EEA producers: 'Indirectly, lower volume from Ilva made the 3 main suppliers (AM, TATA, Thyssenkrupp) more important, increasing suppliers’ negotiation power and inflating product price.' (413)

9.4.3.6. Ilva's impact on prices goes beyond the commodity market

(602)    Despite the fact that Ilva produces primarily commodity steel products, Ilva can indirectly also act as a constraint for the pricing of specialty products to the extent that the pricing of specialty products is related to commodity products (in view of the pricing mechanism described in Section 5.4). (414)

(603)    As explained by a customer, for instance, 'while Ilva is not yet a major Auto player, due to its size, it could still influence the prices in EU for Auto steel'. (415)

9.4.4. EEA suppliers would not be able to and have no incentives to constrain the merged entity

(604)    The Commission considers that EEA competitors would be unlikely to offset any anti-competitive effects stemming from the Transaction. This is for the following reasons, set out in recitals (605) to (622).

(605)    First, as indicated in Section 7.2.5.5, the main EEA flat carbon steel producers are primarily active in Northern Europe. As indicated in the sales shares presented in Table 24, they have a limited presence in Southern Europe, where the main effects of the Transaction would be likely to occur. This is particularly the case for HR, as EEA flat steel producers sell the majority of their HR volumes in the countries where their manufacturing facilities are located.

(606)    Second, based on the Commission's market reconstruction EEA competitors appear  to have limited spare capacity readily available, and hence limited ability to expand their sales. [Internal document].

Figure 74 – ArcelorMittal's internal document […] (Attached to ID1660-41054)

[…]

(607)     […].

(608)    The Notifying Party contested the Commission's spare capacity computation in its Reply to the SO, (416) qualifying it as overly conservative and arguing that the production limitations, identified by the Notifying Party in the reactivation of Tata Steel's plant in Llanwern and in bottlenecks at primary steel production, could be easily lifted. In this respect, the Commission notes the following.

(609)    In the first place, Tata Steel does not appear to have the incentives to reactivate its Llanwern HR mill, as explained in recital (619).

(610)    In the second place, an increase of hot rolling output based on slab insourcing from third parties as claimed by the Notifying Party (417) appears implausible. The Notifying Party claims that there is ample spare slab supply capacity which could feed the existing hot strip mill capacity in the EEA. The Commission notes, however, that only one competitor of the Parties could benefit from such strategy.

(611)    In the third place, [internal document]: [internal document] (418) […]

[internal document] (419) […] [internal document] (420) […] [internal document]. (421)

(612)    […]. Furthermore, it seems that the market for slabs can be illiquid and that an integrated player may prefer to support the needs of its group companies over the demand by other players. This makes it less feasible to expand HR capacity relying upon external slab supplies. […].

(613)    On that basis, the Commission considers that its computation of the spare capacity in the EEA market approximates well the actual ability of EEA competitors to expand following the likely price increase from the merged entity post-Transaction. (422)

(614)    Third, it is unlikely that EEA competitors would increase their own capacity with further investments.

(615)    In the first place, the investments required for a capacity increase which would include an increase in capacity for primary steelmaking facilities, would be significant, as explained in Section 5.1.3.1. In the same vein, greenfield market entry of new competitors is highly unlikely. In the EEA, there have not been investments  in greenfield steel plants since several years.

(616)    In the second place, EEA competitors, including the main competitors of the Parties, generally indicated, in response to the market investigation, that they have no plans  to increase their primary steelmaking and HR capacity. This is also reflected in the behaviour of the last two years. For HR the spread over the raw material costs increased significantly between 2016 and the end of 2017. (423) Despite this general increase in spreads there was neither an increase in steel capacity nor any re- activation of mothballed  plants  as  it  is  shown  in  the  capacity  shares  presented in Section 9.3.

(617)    Fourth, due to the differentiating features of the EEA flat steel market, both from a product and from a geographic perspective, as described in the Horizontal Merger Guidelines, (424) competitors would be unlikely to be incentivised to defeat the merged entity's price increase post-Transaction by undercutting its prices in an attempt to sell more volumes and increase shares. Rather, competitors may increase their prices in response to the merged entity's price increase. This is because part of the demand lost by the unilateral price increase of the merged entity will be recaptured by EEA competitors, who, in turn, will be incentivised to increase prices. By doing so, EEA competitors will benefit from higher sales at higher prices.

(618)    Fifth, EEA flat steel producers' lack of ability and incentives to expand production is demonstrated by the recent increase in HR margins, both in Northern Europe and in Southern Europe. Figure 75 displays the spread of HR in Northern and Southern Europe. The spread is calculated as the difference between HR market price and the basket of raw materials used to produce HR. Hence, this spread can be considered as a proxy of the HR margins. The figure shows that for a prolonged period of time from mid-2016, and in particular during 2017, the HR spread has increased significantly relative to the period 2009–2015. This means that HR margins for EEA steel producers increased. If there was available spare capacity in the market, and if EEA competitors had the incentives to increase production, the increase in HR margins depicted in the figure would have been unlikely, as EEA competitors would have increased their sales and depressed the HR prices in the EEA.

(619)    Sixth, some competitors with idled capacity have not reactivated their capacity even in face of higher margins. For example, it appears that Tata Steel mothballed its HR line in Llanwern (in the UK) in August 2015. (425) From mid-2016 onwards, Tata  Steel's capacity in Llanwern was not reactivated, even though margins opportunities in the market significantly increased.

Figure 75 – HR spread over raw material in Northern and Southern Europe (426)

[Parties’ submissions]

(620)    In its Reply to the SO, (427) the Notifying Party argued that production decisions and  the related investments to expand production are ‘sticky’, and suppliers would not take such decisions if the increase in profitability appeared short-lived. In this  respect, the Commission notes that recitals (614) to (616) refer to capacity that could potentially be brought to the market with limited constraints in terms of investments and business decisions. (428) As such, the argument put forward by the Notifying Party does not apply to those expansion decisions.

(621)    The Notifying Party further argued that EEA suppliers have a sizeable export of HR outside the EEA that could be redirected. Specifically, the Notifying Party estimated that there are […], as shown in Figure 76. The Commission notes that by exporting outside the EEA below European domestic price (which also likely involves higher transport costs than selling in the EEA), EEA suppliers demonstrate strategic behaviour whereby they export outside the EEA in order not to contaminate and lower the EEA price. As such, the Commission considers that there is no reason to believe that those exports are likely to be redirected into the EEA to any significant extent. On the contrary, this evidence shows a low degree of competition among  EEA suppliers and that EEA suppliers have a degree of market power already pre- Transaction, and further indicates that competitors are unlikely to expand sales to defeat likely price increases by the merged entity

Figure 76 – HR Exports below EEA price (429)

[…]

(622)    Seventh, the prevailing prices in Southern Europe are lower than in Northern Europe. EEA competitors located in Northern Europe can therefore already obtain relatively higher margins in their area where they are mostly active. Therefore, any price increase by the merged entity in Southern Europe would have to be very significant before Northern European competitors would have an incentive to shift volumes to Southern Europe.

9.4.5. Imports are not a sufficient constraint to offset the non-coordinated effects of the Transaction

9.4.5.1. The Notifying Party's views

(623)    The Notifying Party considered that non-EEA imports would constrain the merged entity, in particular in relation to commodities in Southern Europe. In support of this consideration, the Notifying Party submitted a number of arguments.

(624)    First, the Notifying Party submitted that the EEA flat carbon steel industry is driven by global dynamics.

(1)   In this respect, the Notifying Party submitted that the intense global steel trade is driven by massive overcapacity at global level. The importance of overcapacity for the EEA steel industry has been acknowledged by industry  and international stakeholders (including the OECD, G20, and the Commission).  (430)

(2)   The Notifying Party also submitted that the global overcapacity will likely remain for the foreseeable future. Although China has pledged to reduce its overcapacity, there is significant uncertainty as to whether, and to what extent, it will do so. (431)

(3)   Additionally, the Notifying Party submitted that there is sufficient global divertible HR capacity, even accounting for trade measures. [Parties' submissions] .(432)

(4)   Furthermore, the Notifying Party submitted that imports of flat steel products into the EEA have been significant and increasing. [Parties' submissions].

(5)   In addition, the Notifying Party submitted that recent HR price-spread  increases in Southern and Northern Europe are not specific to these regions but rather worldwide dynamics shared by other markets like China and Turkey. (433) Recent price increases would not be indicative of a weakening of the import constraint but reflect an increase in raw material costs and a surge in demand in China.

(6)   Moreover, the Notifying Party argued that opportunistic behaviour of non-EEA producers which choose their export destinations according to differences in relative prices make the constraint from imports binding in the EEA and leads to the harmonisation of prices globally. (434)

(625)    Second, according to the Notifying Party the key issue is not the purchasing pattern  of customers in the past but the ability of EEA customers to switch to imports for   the […] EN commodity steel grades that Ilva produces. (435) According to the Notifying Party, all EEA customers, regardless of their size and segment, have effective access to imports, directly or via traders/distributors, in particular for commodity products.

(626)    Third, the Notifying Party submitted that Ilva is affected by the same factors purportedly preventing customers from switching to imports. These factors include long lead times, restricted lot sizes, lack of flexibility and handling of quality issues. ArcelorMittal argued that these factors apply equally to Ilva and are easily remedied through the use of specialised traders. This in particular applies to customers that purchase commodity products.

(627)    Fourth, the Notifying Party also argued that the significance of the import constraint does not depend on volumes at a given moment in time, but the high volumes of imports observed over recent years would indicate that customers can switch to imports so that any constraints about limitations in the import constraint would be irrelevant (436)

(628)    Fifth, the Notifying Party claimed that ArcelorMittal's own internal documents confirm the import constraint.

(629)    Sixth, the Notifying Party submitted that the SO’s case is supported by a flawed econometric analysis that suggests that imports would not materially increase in response to an increase in relative prices, and asserts that imports only constrained EEA producers during the dumping period identified by DG Trade. In this regard, the Notifying Party submitted that a proper analysis of the sensitivity of imports to relative price changes confirm that importers tend to opportunistically react to relative price changes, predicting, on a conservative basis, a […]% HR import increase in response to a […]% relative increase in EEA prices.

(630)    Seventh, the Notifying Party considered that trade defence measures have not reduced the import constraint. (437)

(631)    In the first place, according to the Notifying Party trade measures only diverted imports to non-affected countries. (438) Following the imposition of trade defence measures, non-EEA imports came instead from Turkey, India, South Korea and other countries. The Notifying Party argued that the increase in imports from countries not affected by anti-dumping measures offset the decrease in imports from the affected countries. In addition, ArcelorMittal stated that such 'redirection effect' is reflected in its internal documents.

(632)    In the second place, the Notifying Party submitted that despite trade measures there  is no persistent decrease in non-EEA imports volumes. ArcelorMittal also submitted that, using the most recent data available, which span until June 2017, imports levels in the EU increased by […]% in the first half of 2017 when compared to the first half of 2016 rather than to the last half of 2016 (when a decrease in imports can be observed). The recent decline in import volumes in the second half of 2017 reflects a decrease in the EEA domestic price premium, making the EEA less attractive to importers. Moreover, imports have returned to very high levels in January 2018. (439)

(633)    In the third place, the Notifying Party submitted that non-affected countries have the ability and incentive to further increase imports into the EEA. Non-affected countries that currently supply material volumes to the EEA (India, Turkey, South Korea), have excess capacity in HR that vastly exceeds the current level of HR imports from the countries affected by the anti-dumping measures. According to the Notifying Party, the non-affected countries could further increase their imports into the EEA, even if faced with increased domestic demand. Moreover, Turkish producers  benefit from a free trade agreement with the EEA and given their proximity to the EEA could be considered to form part of the internal market for analytical purposes. (440) Furthermore, many of these countries are now affected by current US trade measures on steel. The Notifying Party submitted that the countries affected by the US measures could redirect the volumes previously shipped to the US to other geographies, including the EEA, further increasing the pressure from imports. (441)

(634)    In the fourth place, the Notifying Party submitted that the countries affected by the EU anti-dumping measures may resume HR imports into the EEA in response to a hypothetical price increase by the merged entity. Thus, it argued that the trade measures did not insulate the EEA from import constraint. As merger assessment is forward looking and trade measures are not merger-specific, any effect on prices which existing measures might have had (quod non), is irrelevant for the assessment of the assessment of the constraint from imports going forward.

(635)    In the fifth place, the Notifying Party considered the US experience with anti- dumping and anti-subsidy orders as an example that shows that trade  measures cannot prevent the influx of imports.

(636)    In the sixth place, [Parties' submissions].

9.4.5.2. The Commission's assessment

(637)    The Commission's investigation in relation to imports of flat carbon steel products makes it clear that while imports represent an important source of supply in the EEA and in particular in the areas where the competitive interaction between the Parties is most prominent, they would be unlikely to be able to sufficiently constrain the merged entity to prevent the non-coordinated effects of the Transaction.

(638)    The Commission's investigation confirms the Notifying Party's claim that imports, taken as a whole, are particularly relevant for the assessment of the Parties' overlap both in terms of the geographic area (Southern Europe and, in particular in Italy) and the products concerned (commodity products).

(639)    However, the appropriate test for the Commission's assessment is not whether imports compete with the Parties. Nor is it, as the Notifying Party correctly points out, whether imports have the potential to replace all of the demand in the EEA. The relevant question for the Commission's assessment of this Transaction is whether imports would exert a sufficient constraint to likely eliminate the incentives for the merged entity to raise prices as a result of the elimination of competition between the Parties as a result of the Transaction.

(640)    The assessment in this Section focuses on the market for HR products, while the assessment of the role of imports with respect to CR and GS is  presented  in  Sections 9.5.5 and 9.7.5 and 9.8.2 respectively.

9.4.5.3. Significant import volumes are not as such an indicator of sufficient price pressure from imports

(641)    In the Commission’s view, the presence of significant import volumes should not be confounded with the existence of sufficient competitive pressure from imports.

(642)    First, at present, significant import volumes appear to be necessary to balance supply and demand in the EEA. In view of the limited spare capacity readily available in the EEA and in light of the export activities by EEA producers imports may therefore be required to complement EEA  producers'  sales  and  fulfil  the  HR  demand  in the EEA, as identified by the Commission's market reconstruction. This is somewhat echoed by market analysts reports, which suggest that 'import sources are essential as domestic mills are not prioritizing commodity grades'. (442) This may explain why relatively significant import volumes remain even after the imposition of the trade defence measures. However, a structural need for significant import volumes does  not imply that imports determine prices in the EEA or parts thereof.

(643)    The Notifying Party correctly pointed out that total EEA capacity is theoretically sufficient to supply EEA demand. However, Europe is a net importer of flat steel products. (443) While redirection of EEA exports to the domestic market might improve the supply and demand balance of the EEA by reducing the need for imports, as explained in recital (621), EEA exports are unlikely to be redirected to the domestic market. Moreover, contrary to the Notifying Party's argument that opportunistic imports by non-EEA producers would imply that non-EEA producers set prices in  the EEA and globally, (444) the coexistence of exports by EEA-producers and imports into the EEA by non-EEA producers indicates that EEA-producers have a degree of market power and strategically choose to sell abroad at a lower price than in the EEA despite the fact that foregone sales in the EEA are taken up by imports. This is shown in Figure 76 and implies that imports are unlikely to fully replace foregone sales in the EEA and that they are not a sufficient constraint to eliminate strategic price increases post-Transaction. If prices in the EEA and globally were purely set by non-EEA producers' actions, there would be no reason for EEA producers to export part of their production rather than selling it domestically.

(644)    Second, periods of high import volumes do not necessarily go hand in hand with strong pricing pressure from imports to the point of depressing domestic spreads in the EEA and driving significant price decreases. For example, during the pre-crisis demand peak (2006–2008), imports were at comparable if not higher level as during the period 2014–2017. However, prices were at the opposite ends of the extreme: while in 2008, prices of HR reached EUR […]/t, they plummeted below EUR […]/t at the end of 2015. (445) Prices in Southern Europe relative to those in China were also significantly higher during the period 2006–2008 than post-crisis (see Figure 77). (446) This indicates that import volumes alone are not indicative of the price level, or that imports would be sufficient to prevent divergence in prices in the EEA relative to other areas. This will depend on a number of additional factors, notably the demand and supply conditions in the exporters' domestic market, in the EEA, as well as in other plausible destination markets for exports, and the cost of raw materials.

(645)    Third, the impact of the surge of imports during the period 2015–2017 should not be assessed without considering the economic and legal context which gave rise to the imposition of trade defence measures by the Union. In particular, in the regulations imposing anti-dumping and/or anti-subsidy duties against imports of certain steel products from China, Brazil, Iran, Russia and Ukraine, the Union considered that the dumped imports from these countries caused unfair trade conditions and led to a material injury to the Union industry. (447) Thus, imports from most of the source countries during the period concerned by the measures deviated from the conditions of normal competition. Therefore, the import levels from the period of dumping are unfit to serve as a measure of the degree of constraint that imports can exert going forward while remaining within the boundaries of the EEA legislation. Moreover, the prices in Southern Europe relative to prices in China do not appear to have decreased during the period 2015–2017 when imports saw a sharp increase (see Figure 77).

9.4.5.4. Non-EEA producers are individually small; they have individually limited influence on prices in the EEA

(646)    While imports may make up a significant share of supplies within the EEA in aggregate, non-EEA producers are fragmented and have, as individual market operators, limited ability to influence prices in the EEA. Imports hence cannot be regarded as actors with a strategic interest in the market but only act opportunistically.

(647)    First, imports represent the aggregate supplies of many, mostly independent producers, and cannot be assessed as if they represented the output of a single entity.

(648)    While the Notifying Party submitted that there is significant steel overcapacity at the global level that can be diverted and imported into the EEA, the Notifying Party was not able to identify the individual exporters, with the exception of a few. While countries like China, Russia, Brazil, or Turkey, among others, are often quoted as important steel exporting countries, the Notifying Party has not provided the Commission with information on all the different steel producers active in those countries and elsewhere and their HR production capacities, except for a few. (448) Likewise, the market share information provided by the Notifying Party also groups all imported HR products into one 'player', that is 'imports into the EEA'. As reliable more granular data has not been made available to the Commission, the computed consumption shares provided in Section 9.3 also group all imports into one category.

(649)    Nevertheless, imports should not be treated as one homogeneous actor in the market with identical ability and incentives to compete as a major supplier. The Commission’s investigation confirms that even imports from a certain country do not come from a single source, but from a number of different steel producers.

(650)    Second, individual exporters, with few exceptions, are likely to be significantly smaller than ArcelorMittal, and very often also smaller than Ilva, as the information provided by the Notifying Party appears to suggest in relation to the manufacturers of flat carbon steel products that it has identified.

(651)    For instance, Erdemir, the largest Turkish steel manufacturer has a current flat finished steel products capacity of c. […] Mt, (449) out of which only […] Mt  is effective HR production capacity. (450) Hesteel, a Chinese group, produces HR and CR and has a total capacity of […] Mt at its plant in Serbia designated for export to the EEA, (451) thus significantly smaller than that of ArcelorMittal or Ilva. Russian  producer Severstal, according to the information provided by the Notifying Party, operates an effective HR capacity of just above […] Mt, (452) similar to JSW Steel's (India) […] Mt. (453) Other manufacturers have a significantly smaller capacity and not all of them produce HR. For example, from those non-EEA producers that the Notifying Party has identified as producers of HR, Colakogluis, a Turkish producer, has a HR production capacity of c. […] Mt. (454)  Habas, another Turkish producer,  has a current HR production capacity of c. […] Mt. (455) Tosyali Holding, also a steel producer based in Turkey has […] Mt HR production capacity. (456)

(652)    In fact, only a very few global players identified by the Notifying Party seem to have HR production capacity larger or comparable to Ilva's. Those include, for instance, Chinese Baosteel, South Korean Posco and the Russia-based MMK. (457)

(653)    In summary, 'imports' represent a multitude of players from different countries and with different commercial strategies, prices, and available divertible capacities.  While the Notifying Party argued that non-EEA suppliers to the EEA market have much higher combined capacity than Ilva, (458) they cannot be considered or assessed  as a single market operator because of their fragmentation. Instead, the role of each non-EEA supplier in the EEA appears to be too limited for them to significantly influence prices individually, and they rather appear to be generally ‘price-followers’ in the absence of exceptional circumstances such as dumping (see recital (645)). This is consistent with comments from a market participant that 'Most of the import [sic] tend to follow the EU producer when they offer into the EU. If EU suppliers rise the prices, then import will quickly react and life [sic] the prices as well, granting only a limited discount with respect to the EU producer.' (459)

(654)    The Notifying Party submitted that the disaggregated nature of importers would rather indicate a stronger constraint from imports as there is a multitude of potential exporters that could increase their sales into the EEA. (460) However, the evidence indicates that there are at least some periods in which non-EEA suppliers are not making competitive offers to EEA customers: ‘South European coil producers are trying to continue increasing their prices incrementally amid strong demand and a lack of third-country imports, sources told S&P Global Platts Friday’. (461) This indicates, as further developed in the following Section 9.4.5.5, that the constraint exerted by non-EEA is not as a reliable source of supply for EEA customers than domestic producers.

9.4.5.5.  Compared to domestic producers, imports are a less reliable source of supply for EEA customers

(655)    Unlike established domestic EEA suppliers, imports do not represent a stable, lasting and dependable source of supply for EEA customers. Imports are characterised by their volatile presence in the EEA and by their opportunistic nature. Therefore, even  if at times imports exercised a more appreciable degree of price pressure, they are not capable of structurally and continuously disciplining the domestic flat carbon steel suppliers in the EEA.

(656)    First, the Commission's market investigation demonstrates that, much like their EEA counterparts, non-EEA flat carbon steel producers tend to focus on their domestic market, which is their core area of sales, and only export certain volumes if the  export market conditions are favourable. In particular, it appears that exporters follow  an  opportunistic  strategy,  whereby their  supplies  to  the  EEA  would  be a function of fluctuations in price and considerations on prevailing conditions of demand in their own domestic markets and other alternative export destinations.

(657)    A number of market participants confirmed that non-EEA producers are unreliable; they simply 'come and go', (462) and focus on their domestic markets first. (463) The non-EEA suppliers themselves confirmed that they tend to focus on their respective domestic markets. A non-EEA producer, which is one of the largest and more widely recognised non-EEA supplier into the EEA, also indicated that about 70–85% of  their production is destined to the domestic market while a small share of 15–30% is allocated for exports (only part of which goes to the EEA). (464) Furthermore, the same producer explained that: 'The Indian domestic market is clearly preferred by the company due to its well established customer base and distribution network, which ensures a sustainable business. Thus, in the event of an increase in the domestic demand, [the company] reallocates the volume of products that would have been exported for the satisfaction of the domestic market’s needs.' (465) Similarly, a Turkish producer identified several differences between non-EEA and EEA-based integrated suppliers: 'Import sources are mostly temporary due to quick changing market conditions. It is difficult to establish a regular business. Also logistically and service wise options are limited. However an integrated mill within territory is reliable and long term partnership is easy to establish together with a lot of service options.' (466)

(658)    This perception is echoed by EEA customers. As a large customer of HR in Italy noted, non-EEA producers' focus is 'first on the domestic market and only then on exports' and for instance Turkey would export to Europe maximum 15-35% of the production. (467) Another customer explained that the importers' presence is merely opportunistic as they primarily focus on their domestic market (Asia) and ultimately sell into the EEA to the most convenient customers. (468)

(659)    As an indication of this lack of incentive, at least in some periods of time, the Commission observes that in the second half of 2017 the level of the HR spread over raw material cost increased by […]% in the North of Europe and by […]% in the South of Europe, compared to the second half of 2016. Over the same period the  level of HR imports in the Union declined by […]% (Figure 78). These events show that there are periods of time during which non-EEA suppliers are not constraining the European prices as an increase in price spreads in Europe did not trigger higher imports but was rather associated with a decline.

(660)    Second, like EEA producers the incentive for non-EEA producers to sell in the EEA depends on the price they can achieve by such sales relative to their sales opportunities in other markets and, in particular, in their home market. However, unlike EEA producers, non-EEA producers have no specific incentives to establish   or maintain a strong market presence in the EEA, which is not their home market. Non-EEA producers may want to benefit from export opportunities when relative prices are favourable or when they face excess supply in their domestic markets. In both cases, the prevailing price levels in candidate export markets will inform the exporter’s selection of the market offering the highest returns. This opportunistic and erratic behaviour of non-EEA suppliers is also reflected in the market assessments produced by the specialised press: ‘There is very little HRC being allocated for  export by Indian mills as their focus is to service the domestic market’ (470) In the first place, an analysis of spreads in various markets will identify where there is scope for arbitrage. Figure 79 illustrates the price spread over raw material prices for HR in various countries, notably the US, China, Turkey, Japan and Europe (South and North).

Figure 79 – HR price spread over raw materials

[…]

(661)    It can be observed from the figure that the Turkish domestic spread is often above the spread in Southern Europe, which suggests that the Turkish exporters may not have a sufficient incentive to sell HR to Southern Europe. This observation seems to be in line with the considerations of Turkish producers. One of them explains: 'As  a general consideration, in this moment the EEA market is not particularly attractive and profitable for Turkish steel operators which prefer to export their products in other countries.' (471) This statement also indicates that export destinations depend on arbitrage opportunities as also indicated by the following statement from the specialised press: ‘Indian mills are offering to Europe these days because the Vietnamese market is too competitive’. (472)

(662)    In the second place, the opportunistic behaviour of non-EEA producers also seems to be supported by historical data, which suggests that the HR import flows are closely related to the difference in domestic prices in the EEA, and in particular in Southern Europe (which is where majority of imported HR into the EEA land), and in the exporting country, such as China (see Figure 77 above). If prices in Europe increase relative to prices in China, imports will increase. However, also the opposite holds true, namely if prices in China increase relative to prices in Europe, imports will decrease.

(663)    Third, the Commission notes that, over a longer timeline, significant fluctuations of import volumes can be observed. For example, the period between 2006–2008 was characterised by high levels of imports and followed by a period when imports of HR decreased and then subsequently increased again in 2011, as depicted in Figure 80 (HR imports into the Union) and Figure 81 (HR imports into Italy).

(664)    Those figures seem to indicate that there are some cyclical patterns in the flows of imported HR. The cyclical patterns suggest that the imports' presence on the market might be volatile and that, even when imports do exert a constraint on the EEA market, such a constraint is transitory and not comparable to the permanent nature  of the constraint imposed by EEA-based steel producers. At any rate, import volumes as such may not allow a conclusion as to the degree to which they exercise price pressure on the EEA incumbents. (474), (475)

(665)    Fourth, the non-EEA importers' opportunistic behaviour is also reflected in the commercial conditions and nature of steel purchases from non-EEA producers. Market participants have confirmed that sourcing of HR from importers is mostly on a spot basis with no long-term commercial agreements. An industrial customer explained in this respect that '[t]he problem with Chinese suppliers is they are too unpredictable mainly in price and do not offer a stable supply but only offer spot sales for 3-4 months. […] The Chinese may also primarily fill their domestic  demands and only supply extras abroad.' (476)

(666)    According to the Notifying Party, one would expect each supplier to supply the bulk of their volumes closer to its plants and incremental volumes further away in an efficient market and that, in light of their available spare capacity non-EEA  producers could easily respond to hypothetical price increases in Europe. (477) The fact that exports account only for a small portion of some exporting countries production would be irrelevant. If anything, for countries like Turkey or India, even a small percentage of their production would amount to significant volumes. Imports from these countries could therefore substantially increase without significantly compromising their domestic consumption. Moreover, the Notifying Party argued that opportunistic behaviour by imports, in the sense that imports react to changes in relative price is the very characteristic that makes imports a binding constraint, and leads to the harmonization of prices globally. (478)

(667)    The Commission agrees that it is natural that suppliers focus on their home markets  as demonstrated by the evidence (set out in this Section) from the market investigation, in line with the observation that exports account only for a small portion of some exporting countries' production.

(668)    However, the argument that imports from exporting countries could increase without significantly compromising domestic consumption in these countries is beside the point, which is that non-EEA producers, who focus on their home markets, are not a reliable and permanent source of supply for EEA customers, as evidenced by the previous behaviour of non-EEA producers. This is also reflected in the commercial conditions offered by importers. (479)

(669)    As regards the constraining effect of opportunistic behaviour of imports, the Commission notes the following.

(670)    First, as noted in recital (621), the fact that EEA producers export significant volumes to non-EEA destinations at a lower price than in the EEA indicates they have a degree of market power as they strategically choose to forego profit opportunities to export volumes rather than sell in the EEA.

(671)    Second, if imports were able to constantly exert such a binding constraint on the  price of EEA suppliers, then HR spread in the EEA (prices minus raw materials) should be aligned across the world (except for transport cost differences). This is not in line with the observation of significant volatility and differences in such spreads as shown in Figure 79 and Figure 82. (480)

(672)    Third, as explained in the following subsections, opportunistic reactions of imports to changes in relative prices is unlikely to sufficiently constrain the merged entity as, in particular:

(1)   The empirical evidence indicates that import reactions to price increases are likely insufficient to defeat price increases (Section 9.4.5.6).

(2)   The ability of imports to exert competitive pressure on EEA producers has  been significantly reduced compared to the period of market distortions in recent years (Section 9.4.5.7).

(3)   Imports  can  only  contest  part  of  the  demand  of  EEA   customers   (Section 9.4.5.8).

9.4.5.6. Empirical evidence indicates that import reactions to price increases are likely insufficient to defeat price increases

(673)    Empirical evidence supports the finding that non-EEA producers, jointly or individually, do not have the ability and incentive to discipline price increases in    the EEA.

(674)    First, as indicated in recital (644) and Figure 77, prices in the EEA and, in particular, relative prices in Southern Europe compared to China, have in the past remained at significantly higher levels than currently for prolonged periods of time, despite an influx of imports similar to the levels reached during the dumping years. This indicates that increases in import volumes that follow increases of prices in the EEA relative to prices in other regions of the world are unlikely to be sufficient to defeat such price increases.

(675)    Second, this conclusion is supported by an econometric analysis of historical data on how import levels are affected by relative prices and other events (which is described in detail in the Annex 1 to this Decision). The analysis indicates that the share of imports in HR sales in Southern Europe would increase by approximately [0-5] percentage points for a 1% increase in the EEA price (that is around [0-5] percentage points for a 5% price increase). Such an increase in the share of imports would be unlikely to be sufficient to defeat a post-Transaction price increase by the merged entity. (481), (482)

(676)    Third, price data (for imports and domestic producers in Southern Europe) depicted  in Figure 82 indicate that while imports are often cheaper than domestic prices, they can  also  be  more  expensive,  as   it   often   happened   over   the   period considered (2009-2017).   Those   developments   suggest   that   non-EEA producers shipping products to the EEA are following independent pricing strategies and do not necessarily undercut domestic EEA producers. This is also apparent from the developments after 2016. Between 2016 and the end of 2017 the spread of domestic prices over raw materials increased significantly. Over the same period also the spread of import prices over raw materials increased by the same magnitude. Hence, over this period both domestic producers and non-EEA suppliers increased their margin on sales to Southern Europe, suggesting a weakening of competition and indicating that imports were not able or willing to constraint EEA domestic producers.

(677)    In the Reply to the Letter of Facts, the Notifying Party claimed that the increase in spreads is a worldwide development and that, therefore, the increase in spreads in Europe would not indicate a worsening of competition in Europe due to a reduced import constraint. In support of this statement, the Notifying Party provided a table showing the increase in the average spreads over raw material for China, Turkey, Southern and Northern Europe. (483)

(678)    In this respect, the Commission considers that the HR spread increases observed in several regions of the world in 2017 are undisputable. However, the extent of the increase was different depending on the region, as shown in Figure 79. For example, in 2017, the HR spread in Turkey increased less than in Southern Europe, despite Turkey being one of the main countries from where imports into Southern Europe originated in 2017. Moreover, in 2016 the HR spread in Turkey was higher than in Southern Europe and in 2017 the spread became lower than in Southern Europe, arguably providing Turkish exporters with even more incentives to sell to Southern Europe. Still, Turkish producers did not increase sales by such an amount as to prevent margins to increase more than in their home market.

(679)    Moreover, even the data submitted by the Notifying Party (Table 5 of the Notifying Party’s Reply to the Letter of Facts) indicate significant variations in the developments of spreads. Between 2016 and 2017 the HR price-spread increased by EUR […]/t in China and by EUR […]/t in Turkey. Over the same period the increase observed in Europe amounted to EUR […]/t and EUR […]/t in the Southern and Northern part of Europe respectively. These variations indicate significant heterogeneity across regions and a significantly higher increase in spreads in Europe.

(680)    Therefore, while an increase in global demand may have contributed to the increase in HR spread worldwide, part of the increase in Europe appears to be due to a weakening of competition which was not counter-balanced by the behaviour of imports.

Figure 82 – HR price spread over raw materials for Southern Europe domestic prices and Southern Europe import prices

[…]

Source: [Parties' submissions]

(681)    The opportunistic behaviour of imports is also shown in some market analyses of the specialised press covering the period from July 2017 to February 2018:

'The trader said he'd not seen offers from India, as that domestic market is strong  and domestic mills are not interested in selling abroad, while Russian mills aren't offering for export either as they're just back from the holidays.' (484)

'The market will probably be more lively as of next week. European producers hiked their prices before the New Year holidays,’ a major trader said. ‘If the market will accept these rises, this will also support Turkish mills' HRC export prices to Europe' (485)

'Imports of hot rolled coil into the European Union fell to the lowest monthly level since December 2014 in October, more than halving compared to a year earlier, according to official data published by Eurofer.

The decline partly reflects the burden of the anti-dumping duty investigation applied to multiple origins, but is mainly attributable to the lack of attractive import offers in the late summer/early autumn seasons.' (486)

Traders have largely struggled to find competitive offers for HRC in the second half of the year as exporting countries have been content to sell domestically or in other markets. Even Turkey and India, the two countries to benefit most from AD duties applied to others, saw year on year declines of 39% and 59%, respectively. (487)

According to sources, Indian import offers have currently been withdrawn from the market as producers are contemplating higher offer prices. Offers from Korea and Brazil were heard at around Eur600/mt CFR Antwerp, up from previous Eur560/mt CFR levels a few weeks ago, making imports less attractive as the spread between imports and domestic levels is tightening to a Eur17/mt difference.' (488)

'Import offers are largely absent in the western European coil market with sources suggesting activity is increasingly limited in the run up to the BlechExpo conference in Stuttgart.

Sources noted even less import options in the market Tuesday, with hot rolled coil supply particularly sparse following the imposition of final anti-dumping duties. A service center buyer in the Benelux said he expects offers to reemerge once it becomes clear what European mills plan to do with Q1 offer levels. “For HRC you hear a lot less offers. India is not in the market. They had to fill up a shipment of CRC a couple of weeks ago, dropped the price, booked a ship and closed their order book,” he said.' (489)

'Third country imports reached €440/mt CIF, with some deals closed at €430/mt CIF Italian ports for September delivery. However, they rose a few euros and now there  is a scarcity of third-country offers, sources agreed'. (490)

9.4.5.7. Ability of imports to exert competitive pressure on EEA producers has been significantly reduced compared to the period of market distortions

(682)    The  trade  measures  against  major  steel  exporting  countries,  described  in  Section 5.3.3.3, appear to remove the market perturbation and thus reduce undue pressure exerted by imports in the recent years of market distortion resulting from dumping prices. It appears that, following the provisional anti-dumping duties adopted by the European Commission  against  Chinese  suppliers  of  HR  in  October 2016, imports of HR into the EEA by Chinese suppliers have essentially ceased while the volumes imported from certain other countries have increased. However, the Commission considers, contrary to the Notifying Party’s claims, (491) that the replacement of imports confirms the effectiveness of the Union trade measures in rectifying the conditions for competition from non-EEA suppliers.

(683)    First, the most recent data show an evident reduction in overall non-EEA import volumes since the adoption of the definitive anti-dumping duties against producers from Brazil, Iran, Russia, and Ukraine in October 2017.

(684)    As depicted in Figure 83 HR imports in 2017 are 18% lower than the HR imports in the same period of 2016. The decreasing trend is recent and seems to coincide with the imposition of further (final) measures in 2017, because most of the volumes loss takes place from September, with the earlier period being instead comparable to figures from previous years. Figure 84 shows that until August 2017 the level of  extra EEA imports of HR was comparable to the level of imports registered over the same period (January–August) in years 2015 and 2016.

(685)    Market analysts have also reported on the significant fall of imports in the EEA as shown in an excerpt from Platts' Global Market Outlook.

(686)    In the December 2017 issue of that publication, Platts qualifies the October 2017's fall of imports of hot rolled coil into the European Union to the lowest monthly level since December 2014 as a 'Steel trade highlight' for Europe. (492)

(687)    Second, the volumes of imports of HR started to significantly decrease already in the period following the imposition of duties against producers from China, and before the last wave of measures in October 2017. This can be observed in Figure 86, which represents the HR import levels on a monthly basis in the EEA.

Figure 86 – HR imports into the EU

[…]

Source: ArcelorMittal's submission, […], slide 21 (493)

(688)    The  total  import  volume  of  HR  has   decreased   by   over   […]%   in   the   period December 2016–April 2017 compared to the previous five months July 2016– November 2016. The figure further shows that HR imports from China more than halved in September 2016 – and ceased starting from October 2016, the month of the imposition  of  the  provisional  anti-dumping  measures  against  China. (494)   Similarly, imports from Serbia, Ukraine, Iran, Russia and Brazil collapsed in December 2016–January 2017, some months after the initiation of the anti-dumping proceeding (July 2016) and simultaneously after the imposition of the registration requirement on Russian and Brazilian imports (January 2017). (495)

(689)    Third, while the import volumes still remain significant, the prices at which  non-EEA producers sell HR in the EEA have increased significantly since mid-2016 (that is after the imposition of the anti-dumping measures), as shown in Figure 87. The same development also applies to domestic prices as shown in Figure 88. (496)

(690)    The figure shows that, since the imposition of trade measures over the last years, the prices have steeply recovered. After bottoming in early 2016 at a level inferior to EUR 300/t, the HR prices rose to above EUR 500/t by mid-2017, and have continued to rise to levels observed before the dumping years and the massive increase of import volumes. To date, specialised press continues to report price increases by  EEA producers:'ArcelorMittal has increased offers for all new coil orders, with hot rolled coil offers in western Europe up Eur20/mt to Eur580/mt ex-works Ruhr, according to sources.The company had withdrawn from the market amid bullish indicators and, with raw material and international steel prices strengthening in recent weeks, the company decided to return to the market with increased offers.' (497) Nearly all of southern Europe's coil producers had come out by Friday with new price increases across product ranges, as expected -- up by Eur20-40/mt, sources confirmed to S&P Global Platts. It's understood the producers aim to incrementally reach Eur550/mt ($667.26/mt) base ex-works for HRC, around Eur640/mt for base ex-works for CRC and for Eur650/mt base ex-works for HDG, with buyers more responsive to price increases.Price are being pushed up by three main factors: a rally in raw materials prices; good demand levels globally; and high import prices with very few offers from outside Europe.' (498)

Figure 88 – European HR domestic prices

[…]

Source: [Parties' submissions]

(691)    Fourth, while price increases may also be driven by cost changes, [Parties' submissions].

(692)    In the first place, the Commission observes that the recent increase in prices has been accompanied by an increase of the HR spread over raw material cost in the EEA and a widening of the difference in the spread of Northern Europe and Southern Europe, as shown in Figure 89. The evolution of spreads shows that the increase in raw material costs does not fully explain the increase in HR prices both in Northern and Southern Europe. The spreads suggest that other factors influence the European prices.

Figure 89 – HR price spread over raw material index

[…]

Note: […]

Source: [Parties' submissions]

(693)    In the second place, this is further corroborated by the evolution of ArcelorMittal's EBITDA as analysed in an internal document in Figure 90.

Figure 90 – ArcelorMittal's Managerial EBITDA overview (499)

[…]

Source: The Notifying Party's reply to RFI 7, Annex 66.1

(694)    In the period from early 2016 to date, […]. (695)       Fifth, […].

(696)     In the first place, […]. (500)

(697) In the second place, […]. (501) Final anti-dumping measures against Russia and Ukraine were imposed in October 2017.

(698)     In the third place, […]. (502) (699)  In the fourth place, […]. (503)

Figure 91 – ArcelorMittal internal document

[…]

(700) [Internal document].

(701)  Sixth, the effects of the anti-dumping measures, namely decrease in volumes, unavailability of certain sources, and increased prices by both EEA and non-EEA producers have also been confirmed by the feedback from customers obtained during the market investigation. (504) The customers indicate that to the extent imports from  the countries affected by the anti-dumping duties have been replaced by imports  from countries not affected by the anti-dumping measures, the HR imported from the non-affected countries has been sold to the European customers at  prices significantly higher compared to the prices that were offered by the countries  affected by the anti-dumping measures. Any substitution of imports from other non-EEA countries was no longer reflecting unduly low dumping prices, indicating that the constraint exerted by alternative sources of imports on pricing acts at a different level than dumped imports.

(702) In the first place, many operators, comprising both suppliers and customers, assert a causal link between higher prices and the trade defence measures. (505)

(703) A flat carbon steel producer indicated that: 'retroactive anti-dumping measures have led to an increase in the price of steel.' (506) Another customer also stated that the anti- dumping measures led to stabilisation and higher prices. (507)

(704) In the Commission's market investigation, a clear majority of customers confirmed the increase in the prices of non-EEA imports. (508) Some customers have further explained that the increase in price was at least partially due to the trade defence measures. (509) A customer active in distribution indicated that post anti-dumping measures, importers from non-affected countries have started increasing their prices to align them with the European prices, even if international prices were lower: 'They have startly increased their prices till european prices, even if international prices were  lower'. (510)  When  asked  whether  the  anti-dumping  measures  have    affected commercial policy of other importers, a majority of customers replied positively. (511)  A customer stated that they 'see higher price level both from EEA and non EEA suppliers'. (512) That customer has stopped all sourcing from Russia and does not evaluate China anymore either. (513) A similar observation was made by another customer that noted: 'reduction of imported volumes, less pressure on market price, improved (healthier) playing field between competitors'. (514) Such  observation suggests that the dumping years indeed caused distortion on the market and the reduced volumes of imports flowing into the EEA after the imposition of anti- dumping measures exert less pressure on the market price, in line with the Commission's assessment of imports as 'price takers'. A German customer also expressed their view that 'the price-level increased dramatically after  implementation of the anti-dumping duties.' (515)

(705) In the second place, the Commission notes that the view of respondents to the market investigation, particularly customers, as regards possible future implications of the trade measures, is aligned with the recent downward trend of import volumes as described in recital (683). This is in contrast with the picture suggested by the Notifying Party's interpretation of the import replacement effects.

(706)  In particular, the replies by customers depict a significant reduction in the pressure from imports following the trade defence measures. Their replies however generally do not consider the injuries on the EEA producers from dumping prices of imports. A trader active in the EEA has also confirmed that the implementation of the trade defence measures consistently reduced the flows of imports from the countries affected by trade defence measures, which has advantaged the EEA producers. For that trader, it has become more difficult to find suppliers located in countries not affected by the measures. (516)

(707) In essence, while before the trade defence measures of 2016 and 2017 a clear majority of both industrial and SSC customers considered the competitive pressure exerted by imports to be strong or at least moderate, following the measures they consider the pressure to be limited or at most moderate. The results are even more telling if considering customers in Southern Europe, and Italy in particular. The majority of both industrial and distributor customers in that region consider the competitive pressure from imports to  be limited  after the trade  defence  measures  of 2016 and 2017. Table 30 illustrates the customers' views at the EEA level on the competitive pressure exerted by imports pre and post the imposition of trade defence measures. Table 31 illustrates the responses to the same questions, but considering only customers located in Southern Europe.

(708)  In the third place, customers that responded to the Commission's investigation state that there is limited availability on the supply side following the introduction of the anti-dumping measures. A customer explained: 'Anti-dumping policies have affected other non EU mills. Less offer and when available the prices were adjusted to the reduced competition.' (517)

(709) Another customer noted that 'China and Russia importers have almost disappeared from steel market in  EEA' following the introduction of anti-dumping measures. (518) Another customer stated that 'import has totally stopped' after the imposition of anti- dumping duties on HR from China. (519)

(710)  An industrial customer observed that following the anti-dumping measures, 'other countries can be afraid to be charged by duties too and so withdraw offers from the market. It reduces sourcing opportunities.' (520) While it is noted that the Commission applies a 'lesser duty rule' when imposing trade defence measures, that is it only imposes duties that are necessary to bring prices back to a normal profitability levels, the same customer also observed that the anti-dumping measures have made it 'more difficult to source from EEA producers because of the price increase they apply (significantly above the cost impact of coking coal and iron ore increase)'. (521) Such observation would be in line with the Commission's finding that imports' ability to exert pressure on prices in the EEA is limited if one disregards the disruptive environment caused by dumping.

(711)  A shortage of HR available for imports has been lamented by certain customers. These include Marcegaglia, the largest consumer of HR in Europe, which due to an alleged lack of availability of HR indicated that it had to import semi-finished slabs and have them re-rolled by a domestic manufacturer (Ilva). While before anti- dumping measures, Marcegaglia sourced 25–45% of HR from imports, afterwards  the volumes of HR imports decreased significantly to 15–35%. The 5–25% shifted to slab purchasing for slab conversion. (522),  (523)

(712) Seventh, the effects of the anti-dumping measures on the market observed by market participants is in line with the Commission's assessment in its most recent anti- dumping regulation on HR where the Commission concluded that while ‘the Commission rejected the claim that the imposition of measures would lead to a shortage of supply of the product concerned/like product in the Union market’, 'the imposition of anti-dumping duties is most likely to lead to a higher dependence of the users on the Union industry.' (524)

(713)  Eighth, the impact of the anti-dumping measures might extend beyond the countries directly affected by the duties. There are indeed signs of a wider deterrence effect by which non-EEA suppliers are becoming conscious of not offering too much volume to Europe as not to trigger possible anti-dumping investigations on them: ‘Most of  the non-European mills are scared of exporting too much into Europe because they don’t want the European Commission to open anti-dumping files on them’ (525)

(714)  Ninth, the Notifying Party submitted that while in the second half of 2017 imports decreased, in January 2018 this trend was reversed indicating that imports flows are not affected by antidumping duties. (526) In this respect the Commission considers the following. In the first place, the increase in imports in January 2018 might be an isolated episode following several months of significantly low levels of imports. Indeed the latest data relative to February 2018 still indicate relatively low levels of imports. In the second place, the relative increase in imports in January 2018 is unlikely to  be  driven  by  the  contemporaneous  increase  in  the  price  premium of European domestic prices over import prices. Indeed, January 2018 import volumes are not traded at the prevailing January 2018 prices, but reflect the outcome of sales closed 3 to 4 months earlier (as the delivery time for imports is at least 3 months). The 3 to 4 months preceding January 2018 were characterised by a negative premium of domestic prices over import prices in the EEA. Therefore, it is unlikely that the January 2018 volumes could be explained as a reaction to a relative price increase (see Figure 92).

(715) Tenth, the Commission agrees with the Notifying Party that the impact of trade defence measures is not merger-specific and that merger assessment is a forward looking assessment. Contrary to the Notifying Party, however, the Commission considers that trade measures are unlikely to only have a one-off effect on prices.  The Commission also considers that such trade measures, by removing a perturbation and restoring fair trading conditions, are likely to weaken the constraint from imports going forward compared to the constraint that imports exerted in recent years. Moreover, the current trade measures are likely to remain in place at least over the forward looking time horizon relevant for the forward-looking assessment of a merger.

9.4.5.8. Imports can only contest a part of the demand of EEA customers

(716)   The ability of imports to reliably and continuously serve the EEA markets is not only limited by their primary focus on domestic sales, but also by the characteristics of the demand by EEA customers. The Commission's investigation indicates that only a limited portion of HR demand in the EEA could be challenged by imports.

(717)  First, both customers and competitors who participated in the market investigation consider that non-EEA imports can only substitute a limited amount of EEA supply for flat carbon steels. (527) While market participants consider that non-commodity grades are more difficult to replace with non-EEA imports, substitution appears limited even for commodity grades.

(718)  Competitors largely acknowledge that their supplies to European customers cannot  be fully replaced by imports from non-EEA countries. (528)

(719)  A majority of EEA customers themselves, while indicating that imports are a source of supply, suggest that they could not switch to imports for a large part of their demand needs, including in respect of commodity grades.

(720)  Feedback from the Commission's market investigation shows that some customers, accounting for over 20% of industrial customers and over 10% of distributor customers, even consider that none or only up to 5% of their demands for flat carbon steels could be satisfied by imports. (529) This suggests that imports may be able to contest and be a viable alternative only for a limited part of the demand for flat carbon steels in the EEA.

(721)  Feedback from SSCs and distributors replying to the Commission's investigation suggests that for  a  majority  of  them  the  maximum  percentage  of  HR  commodity products that they can realistically source from non-EEA suppliers is between 10–40% of their needs. (530) Also, a majority of industrial customers that responded to the Commission's investigation indicated that the HR commodity purchases that they could realistically source from non-EEA suppliers would be capped, mostly indicating 10-40% as the realistic percentage. (531)

(722) Second, even during the dumping period, when some imports were sold at exceptionally low prices, customers were willing to satisfy only a limited portion of their demand with imports. This appears to be empirically consistent with purchasing patterns during the period before the Commission adopted anti-dumping measures (the 'dumping period', going from January 2015, starting period for the anti-dumping investigation against China, until October 2016/end of 2016, when preliminary anti- dumping measures against China and registration measures against Brazil and Russia were imposed). Even during this 'dumping period' the sales share of EEA demand for HR that the importers were able to gain did not rise above 22.2% in the EEA  (or 36.4% for Southern Europe).

(723)  Third, [Parties' submissions].

Figure 93 – Share of HR sales in the EEA by non-EEA suppliers

[…]

Source: [Parties' submissions]

(724)  In the first place, on the basis of Figure 93, it is the Commission's view that in 2015 and 2016, both years when import levels into the EEA were at their peak due to dumping practices, a large majority of customers sourced […] from imports.

(725) In the second place, […].

(726) [Commercial information].

(727)   The analysis refers to 2016, which was a year characterised by dumping prices for   HR imports and very high volumes of imports. [Parties' submissions, commercial information], (532) […].

(728) [Commercial information], the data suggest that the EEA customers that in 2016 purchased between 0 and 10% of their HR needs from imports represent more than […]% of the total number of EEA HR customers, and […]% in terms of total HR volumes purchased (most of the customers in this group did not purchase any volume from imports both in 2015 and 2016). Moreover, customers that purchase less than […]% from imports account for […]% of the total customer number and for […]%   of the total HR volumes sold in the EEA in 2016. The estimations are provided in Table 32. (534)

(729) Moreover, […], the Commission has also analysed the import pattern of only those customers that purchase HR from Ilva. […]. The purpose of this analysis was to focus on customers that source from Ilva and exclude customers that for  some reasons do not source from Ilva (for example because they only purchase specialty products and hence might not consider Ilva nor imports as a possible supply source). The shares of those customers' purchases that are represented by imports are illustrated in Table 34 […] and in Table 35 […]. Again, they suggest that even if  only customers that buy from Ilva are considered, the recourse to imports of those customers is still limited.

(730)   In the fourth place the above analysis, in recitals (724) to (729), [internal document]   is also confirmed by ArcelorMittal's internal documents. […]. (535)

(731)     […]. (536)

(732)     […]. (537)

(733)   However, the Commission observes […] (538)  […]. (539) […].

(734)   Therefore, the Commission concludes that the evidence related to […] together with the other evidence related to other market participants or more generally the market, discussed in recitals (716) to (729), indicate a limited recourse to imports by a significant proportion of EEA HR consumers.

(735)   Fourth, the Notifying Party argued that there is 'no basis to allege that there is a natural limit [customer's ability to rely on imports]' as there would be 'many customers that source significant volumes from imports up to 100%'. (540) Moreover, EEA producers slashed prices in response to imports including to customers not purchasing from imports. (541)

(736)   The  Commission  notes  that  prices  have  indeed  declined  sharply,  in  particular  in 2015, and bottomed out in early 2016 when the Commission initiated an anti- dumping proceeding concerning HR from China and have since recovered strongly  as shown in Figure 88. However, the spread of HR prices over raw material cost has shown a more stable evolution as shown in Figure 89. The spread over raw materials first remained stable in the first half of 2015 it then decreased slightly in    the second half of 2015 and more significantly in the second part of 2016. Both those drops  were much less pronounced than the price decline and after the second half of 2016 the spread kept increasing consistently. This indicates that, while EEA producers lost significant volumes to imports during the dumping period, their price reactions have been more limited and delayed and the overall price decrease during the period was likely largely due to reductions in raw material costs. Given this ability to maintain the price spread and actually to increase the spread over time, the Commission therefore considers that the hypothesis that EEA producers sacrificed their margin to keep customers is not supported by the available data on price spreads.

(737)   Fifth, a number of reasons may explain why customers have limitations as to how much of their HR requirements they can realistically source from imports. In addition to price, which is generally the driver to resort to imported HR, there are a number of other commercial, non-price related considerations, such as longer lead times, logistical challenges, unsatisfactory or lacking service, unfavourable payment terms or commercial conditions, issues with post-delivery claims or the requirements of certain lot sizes that affect customers' procurement decisions.

(738)   In the first place, feedback received from HR supplier shows that customers do consider non-price factors in their purchasing decisions, which limits the customers' willingness to source from imports. A non-EEA producer explained that 'ease on logistics, lead time and service possibilities makes domestic/EEA suppliers more preferable' option than non-EEA producers for customers' flat carbon steel purchases. (542) A trader active in the Union also confirmed that their customers would not buy exclusively imported materials due to reasons such as delivery time, flexibility in terms of quantities that can be purchased, and the customer service. (543) Another, another market participant indicated that customers in the EEA have limitations, due to a number of commercial considerations and non-price factors, as  to how much of their demand can be sourced from imports: 'The high concentration of the EEA steel market coupled with a limited substitutability in terms of quality between EEA and non-EEA steel products leads the Company to be concerned about the possibility to secure volumes at reasonable terms. According to the Company, substitutability between local products and imports is also constrained by: (a) Imports requiring that the orders are placed long in advance, requiring the planning of purchases and forecasting consumption more in advance than when sourcing from EEA suppliers. This is challenging when the Company does not always know the exact volumes of products required by its own customers. Eventually, it may also require problematic and potentially costly changes to orders that may have been placed 3 or more months earlier; (b) Long lead times associated with imports. When shipping from China, the lead times amount roughly to at least five 5 weeks; (c) The very limited distribution services offered in Europe, especially by Chinese suppliers. This factor is relevant due to the fact that the Company does not purchase directly from the mills but rather from distributors, which can perform cutting and stocking services on its behalf.' (544)

(739)   In the second place, customers responding to the Commission's market investigation have  explained  that  non-price  related  considerations  that  pose  difficulties  with sourcing from imports limit the possibility of switching to non-EEA imports. (545) For example, a customer explained that imports cannot substitute domestic/EEA producers due to (i) quality of products; (ii) structure of the supplier-customer relations; and (iii) lead times. (546)

(740)   A clear majority of customers (SSCs as well as industrial customers) have indicated that non-EEA suppliers perform less well than EEA suppliers in particular when it comes to lead times. (547) However, also other factors such as minimum required lot sizes, logistical challenges, payment terms (for example letters of credit), contractual issues and insufficient flexibility are reasons that make sourcing from outside the EEA more difficult or prevent it altogether. (548) One customer elaborated: 'We do [not] buy any HR volumes from non-EEA suppliers because of our product portfolio, flexibility and required end-user / customer acceptance.' (549)) While the minimum commercially viable lot sizes vary and depend on several factors, one distributor explained that the minimum viable size is 5 kilotonnes because 'for less volumes the traders and non EU mills are not interest[ed] in assuming all the administration and logistic costs.' (550)

(741)   One customer explained: 'In term of quality/grades we are in position to largely substitute. In term of delivery time, financial surface necessary to support the purchase and "customized" delivery, only a small amount of purchases could be substituted.' (551)

(742)   Lead times were most often singled out as a barrier to importing (more) volumes from non-EEA producers. (552) On the one hand, long lead times means less flexibility and inability to offer just-in-time delivery services. An industrial customer clarifies: ‘Imports make for a small part of [name of the customer]'s total supplies ([<5]).  That is mainly because of logistics, i.e. lack of flexibility in the supply chain of non-EEA suppliers.’ (553) A customer explained that the company purchases only a minimum quantity (not more than 10%) from non-EEA suppliers. They prefer to be supplied by domestic producers since they can ensure security of supply. (554) Due to long lead times, the price fluctuations may further affect the commercial attractiveness of the purchase. A distributor customer explains: ‘[N]on-EEA based suppliers have very long lead times which generate a risk of high volatility in  prices. [Name of the customer] does not want to risk that during long lead-times prices will go down.’ (555)

(743)   Furthermore, delivery distances do not only entail high lead times but are also associated with other commercial risks. The feedback received from market participants clearly indicates that sourcing from importers comes with a higher level of risk, whether it is risk of delayed deliveries, issues with returns, or currency risks.

(744)   Customers also pointed to a number of other aspects rendering imports less competitive, notably the quality of related services (payment (556), contractual issues, after sales claims (557)) and overall lack of flexibility (for example size of lots).

(745)   Some of these non-price related factors were also identified by customers as a reason why they did not source more HR from imports during the dumping period, even though prices were particularly attractive: 'Financing and excessive stock positions cannot be sustainable at a certain point.' or '…maintaining an European supplying base is strategic for [name of the company].' (558) Another customer, referring to one non-EEA producer, also highlighted the imports' disadvantages that prevent them from sourcing more from imports: 'We have logistic constraints which prevent us from buying more from Zelezara. Indeed they can propose only spot deals and uncertain delivery dates.' (559) Also, an Italian customer mentioned 'Storage not possible, financial constraints, price risk' as limitations to sourcing more from imports. (560)

(746)   The Notifying Party argued that any perceived factors limiting  customers' willingness or ability to switch to imports would also apply to Ilva and that those customers that currently purchase from Ilva should hence be able to credibly threaten to switch to imports. (561) In this respect the Commission notes that Ilva does not currently suffer from the same weaknesses as imports. For what concerns lead times, which is mentioned as one key limitation of imports, Ilva is not at a disadvantage as compared to other producers like ArcelorMittal. Similarly, there is no universal agreement that its service levels are bad. (562) Furthermore, the arguments relating to lead times do not affect the finding of a structural difference between Ilva and imports, due to Ilva's structural availability of significant manufacturing capacity within the relevant market and its incentive to mostly compete for customers within that market. These findings are also without prejudice to the consideration that, even if there was a deterioration of service level, this would likely not have persisted in  the hypothesis of an acquisition by an alternative buyer.

9.4.5.9. EEA suppliers are, to a certain extent, also gatekeepers for imports (747)      [Commercial information]. (563)

(748)     [Internal document].

(749)     In the first place, [internal document]. (750)      In the second place, […]. (564)

9.4.6. The Transaction would remove Ilva as an important competitive constraint  and would likely allow the merged entity to increase prices

(751)    Removing Ilva as an independent competitor with significant capacity and absorbing its capacity would likely significantly strengthen ArcelorMittal’s pricing power in line with the company’s own pricing models, and thus would constitute a significant impediment of effective competition through non-coordinated effects arising from  the removal of an important competitive constraint.

(752)    First, already pre-Transaction, ArcelorMittal is the clear market leader in the EEA, exercising market power due to its unparalleled primary and secondary steelmaking capacity in the EEA. A combination of its capacity and product assortment qualify ArcelorMittal as an important, and in some cases, unavoidable trading partner for flat carbon steel customers. […]. (565)

(753)    In addition, the Transaction would further enlarge the capacity gap between the merged entity ([40-50]%) and its nearest competitors, ThyssenKrupp ([10-20]%) and Tata Steel ([10-20]%), (566) and make it even harder for customers to fully switch away from the merged entity’s supplies, particularly in Southern Europe.

(754)    Second, despite recent difficulties, Ilva remains an important competitive constraint in the EEA, and is a close competitor of ArcelorMittal. [Internal document]. (567) […]. (568)

(755)    Third, by removing Ilva as an independent competitor and adding up its capacity to ArcelorMittal’s, the latter could raise prices more successfully as customers would have even fewer alternative suppliers with limited available capacity to switch to in order to avoid a price increase, in particular in Southern Europe. This applies to both EEA suppliers and imports, which exercise significantly less downward pressure on prices compared to unduly low dumping price levels prior to the introduction of trade defence measures. A larger share of production and supply that cannot be contested by other sources would thus translate into an increased pricing power of the merged entity.

(756)    [Commercial information].

(757)     […].

(758)     […].

Figure 94 – […] (569)

[…]

(759)     [Internal document]. (570)

(760)     […]. (571)

(761)     […].

Figure 95 – […]

[…]

(762)     […]. (572)

(763)     […]. (573)

(764)              Fifth, […].

(765)              In the first place, ArcelorMittal’ strategy documents […].

(766)              In the second place, a large majority of customers located in particular in Southern Europe have considered that the Transaction would lead to price increases, in particular for commodity products in the EEA. Many customers explained that ArcelorMittal already has a degree of market power, which would be further strengthened by acquiring Ilva, one of the main low price suppliers in the EEA.

(767)              Sixth, the Notifying Party arguments are unconvincing. (574) […]. Concerning the first argument, the Commission notes that the different evidentiary pieces consistently identify market consolidation and geographic closeness/natural areas of competitors as some of the key factors contributing to the existence of pricing power. […]. Concerning the second argument, […], this in itself offers no guarantee that the overall output of the ArcelorMittal group in the EEA would be increased. At any  rate, a further increase in the capacity of a market leader rarely translates into a reduction of its market or pricing power. (575)

9.4.7. Conclusion on horizontal non-coordinated effects in the market for the production and supply of HR flat carbon steel products

(768)              With regard to the market for the production and supply of HR flat carbon steel products in the EEA, the Commission finds that the Transaction would strengthen ArcelorMittal’s market position, both in terms of sales shares and, structurally in terms of HR capacity within the EEA.

(769)              The Transaction would eliminate the competitive interaction between ArcelorMittal and Ilva, who compete very closely in Southern Europe and Italy in particular.

(770)              The Commission also finds that the constraint exerted by Ilva on ArcelorMittal is significant, as Ilva has acted as an important competitive force in view of its size and strategy.

(771)              Based on evidence gathered during the market investigation, the Commission also finds that the remaining EEA suppliers would not have the incentive and ability to constrain the merged entity. Similarly, imports would not be likely to sufficiently constrain the merged entity to avoid the likely effects of the Transaction on customers.

(772)              In conclusion, based on the information provided by the Notifying Party and on the results of the market investigation, the Commission considers that the Transaction would be likely to significantly impede effective competition through the removal of an important competitive constraint in the market for the production and supply of HR flat carbon steel products in the EEA as a result of horizontal non-coordinated effects.

9.5. Production and supply of cold-rolled flat carbon steel products (CR): horizontal non-coordinated effects

9.5.1.  Post-Transaction the merged entity would be the largest CR manufacturer, with increased market power in the EEA

9.5.1.1. The Notifying Party's views

(773)              The Notifying Party argued that the Parties' combined market shares  for CR on      the EEA are below the traditional threshold for concern. Those combined market shares do not show market power, particularly given Ilva's commercial collapse and continued import pressure. If anything, the Parties' combined market shares overstate the actual degree of competition between ArcelorMittal and Ilva.

(774)              Secondly, the Notifying Party argued that the only overlap is in the production and supply of commodity products in Southern Europe ,(576) where large amounts of imported commodities would continue to land. Furthermore, the Notifying Party submitted that Ilva principally competes against commodity imports and given its quality, delivery and service issues, is struggling to achieve prices above import levels.

(775)               Third, the Notifying Party submitted that ‘several strong EEA competitors  with excess capacity will constrain the merged entity post-Transaction.’:

(1)   the Parties will continue to face strong competition from several major EEA manufacturers    with     considerable     EEA     sales     shares,     including Tata Steel ([5-10]%), ThyssenKrupp ([5-10]%), VoestAlpine ([5-10]%), US Steel ([0-5]%), SSAB-Ruukki ([0-5]%), Marcegaglia ([0-5]%), Salzgitter ([0-5]%) and ISD ([0-5]%). (577)

(2)   the   Parties'   EEA   competitors   had   'substantial'   capacities    in    2016: Tata Steel ([10-20]%), ThyssenKrupp ([10-20]%), VoestAlpine ([5-10]%), Marcegaglia ([0-5]%), SSAB-Ruukki  ([0-5]%),  Salzgitter  ([0-5]%),  US  Steel ([0-5]%) and Arvedi ([0-5]%).

(3)   several of the Parties competitors had significant excess capacities in 2016: Arvedi  c.[60-70]%,  ISD  c.[50-60]%,  NLMK   Europe c.[50-60]%,   Salzgitter c.[40-50]%, Tata Steel c.[30-40]%, SSAB-Ruuki c.[20-30]%, VoestAlpine c.[20-30]%, ThyssenKrupp c.[20-30]% and Lusosider  c.[20-30]%.

(776)               Fourth, according to the Notifying Party, many of their EEA competitors have [Parties' submissions]. In this regard, it is submitted that ThyssenKrupp, VoestAlpine and Marcegaglia have […] and ThyssenKrupp, VoestAlpine and Tata Steel have  high capacity dedicated to the automotive sector. Furthermore, Salzgitter's […].

9.5.1.2. The Commission's assessment

(777)               Pre-Transaction, ArcelorMittal is already the market leader in the EEA, as well as in Southern Europe.

(778)               At the EEA-level, ArcelorMittal is the largest CR player in terms of sales volume, with a 2016 sales share of [20-30]%. When combined with Ilva ([5-10]%), the  Parties would have an even higher combined share of [30-40]%. Even based on data for 2015, when the Parties had a lower combined market share of [30-40]%, the merged entity would be more than three times larger than the second or third players, ThyssenKrupp ([5-10]%) and Tata Steel ([5-10]%). It would be more than […] times larger than the fourth player, VoestAlpine ([5-10]%).

(779)               The Parties and their competitors’ market shares are presented in Table 36.

(780)               Both ArcelorMittal and Ilva are active in the production of CR products in the EEA. The merged entity would be the producer with the largest capacity in the EEA, as shown in Table 37. Specifically, ArcelorMittal, with a capacity share of [30-40]% (Scenario 1) – [30-40]% (Scenario 2), (579) when combined with Ilva with a capacity share of [5-10]% (Scenario 1) – [5-10].% (Scenario 2) would achieve a combined

capacity share of [40-50]% (Scenario 1) – [40-50]% (Scenario 2).

(781)               The Commission's assessment as to whether it is appropriate to use capacity shares as a relevant metric has already been outlined at Section 8.2.2 and applies equally to the competitive assessment of CR.

(782)               The Parties’ share of capacity in the EEA for CR is shown in Table 37.

(783)               In terms of consumption share, the market is already concentrated with the three major players, (ArcelorMittal, ThyssenKrupp and Tata Steel) accounting for more than […]% of the share of the market. Table 38 reporting the consumption shares at EEA level for the period 2012–2016, shows that ArcelorMittal is already the largest CR producer in the EEA in terms of consumption shares, serving [30-40]% of EEA CR demand. When combined with Ilva ([5-10]%), the Parties would have an even higher combined consumption share of [40-50]% for 2016. The merged entity would therefore be more than […] times larger than the second or third players, ThyssenKrupp ([10-20]%) and Tata Steel ([10-20]%). It would be more than […] times larger than the fourth player, VoestAlpine ([5-10]%).

(784)               The Commission's assessment as to whether it is appropriate to use consumption shares as a relevant metric has been outlined in Section 8.2.2 and it applies equally to the competitive assessment of CR.

(785)               Table 38 provides the Parties'  market  share  of  CR  in  terms  of  consumption  in the EEA.

(786)              As a result of the Transaction, the merged entity would thus significantly strengthen its position as the market leader in CR on the EEA market, whether in terms of sales, capacity or consumption shares. Regardless of the market share category, given the highly asymmetric nature of the market, the merged entity would be more than […] times larger than the next player on the market.

(787)              The Commission's assessment as to whether it is appropriate to look at the impact of the Transaction in Southern Europe as in Section 7.2.5 (580) applies equally to the competitive assessment of CR.

(788)              Concerning Southern Europe, the Parties’ market shares are even higher than in the EEA as whole. In Southern Europe (and in particular in Italy), the merged entity's market position for sales in CR would be sizeable and there would be a strong asymmetry among the market players. The combined sales shares for 2016 would    be [40-50]% with an increment of [10-20]%.The merged entity would therefore be approximately […] times larger than the  second  player  on  the  market,  Marcegaglia ([10-20]%) and […] times larger than the third, VoestAlpine ([5-10]%).

(789)               In Italy, the merged entity's market position for sales in CR would be slightly lower. In this case, combined shares for 2016 would be [30-40]%, with an increment of [10-20]%, represented by the sales shares of ArcelorMittal. The second largest competitor on this market is Marcegaglia with [20-30]%, and thereafter VoestAlpine with circa [10-20]%. Any remaining competitors have very minor shares on the market for CR in Italy.

(790)               The Parties’ and their competitors’ market shares in Southern Europe and in Italy and Spain are shown in Table 39.

(791)               As regards the consumption shares in Southern Europe, Table 40 demonstrates that the Parties' combined shares are substantially higher compared to EEA level shares. In 2016, the merged entity's consumption share was [60-70]% (+ [10-20] relative to EEA level). Post-Transaction, the merged entity would therefore account for the vast majority of consumption of CR in Southern Europe, being […] times larger than the next EEA player, Marcegaglia ([10-20]%) and […] times larger than the third player in the region, Arvedi ([0-5]%).

(792)              Therefore, whilst ArcelorMittal is already the market leader in the industry prior to the Transaction, the merged entity would have an even stronger market lead in CR post-Transaction, however measured. The remaining competitors would be significantly smaller in all respects – the merged entity would be approximately […] or […] times the size of the next competitor.

(793)              To conclude, post-Transaction the merged entity would enjoy increased market power in CR in the EEA as well as in Southern Europe and Italy. This contrasts with the significantly weaker position of the Parties' competitors.

9.5.2.   ArcelorMittal and Ilva are close competitors

9.5.2.1.  The Notifying Party's views

(794)              The Notifying Party submitted that the parties are not close competitors. It argues  that whereas ArcelorMittal's business focuses on the production and supply of high value products to the automotive sector and other demanding customer groups, Ilva's business is more focussed on the production and supply of commodity products. Moreover, […] of Ilva's CR supplies are in Italy and subject to intense pressure from imports and other EEA producers.

(795)               Second, the Notifying Party submitted that ArcelorMittal has no material merchant CR production in Piombino and that ArcelorMittal Florange is the nearest merchant production plant to Italy. It is argued that in 2016, ArcelorMittal Piombino sold only […] of CR to third parties and that Marcegaglia and Arvedi are closer to Ilva's plants. (582)

Source: […]

[…]

(796)              Third, the Notifying Party submitted that there are low levels of EEA customers switching from ArcelorMittal to Ilva. It argued that switching from ArcelorMittal to Ilva 'appears to be very low', in terms of number of customers and in terms of volume: Ilva gained only […]% of EEA customers even when Arcelor lost significantly. The conclusions would be similar for a switching analysis when considering customers in Southern Europe.

(797)              Fourth, the Notifying Party submitted that Ilva's market share in CR has fallen, as capacity constraints at blast furnace level also affect CR.

(798)              [Parties' submissions].

(799)              In its Reply to the SO, the Notifying Party reiterated its arguments to the effect that the Parties are not close competitors, which have been outlined in Section 9.4.2, also in relation to CR. More particularly, ArcelorMittal disagreed with the claim that it is a particularly close competitor of Ilva in basic commodity grades. In this regard, ArcelorMittal argued that the Parties' combined supply shares overstate the degree of their competitive interaction, including in Southern Europe.

9.5.2.2. The Commission’s assessment

(800)              The general arguments of the Notifying Party, which apply across products, and to HR and CR in particular, and which have already been addressed in Section 9.4.2, will not be addressed again in this Section. Reference is thus made to Section 9.4.2   as regards arguments of a general nature (further specific references will be presented in the assessment). In this Section the Commission will instead address those arguments which are specific to the production and supply of CR products.

9.5.2.3. The Parties’ CR geographic footprints largely overlap

(801)              In line with the discussion on geographic closeness for HR at Section 9.4.2.4, it is noted that the Parties compete closely in view of the location of their CR manufacturing facilities. Ilva has a cold rolling mill to serve customers (for instance packaging, automotive, white goods) at Novi Ligure, North West Italy ([…]), Genova, North West Italy ([…]) and Taranto, in the south ([…]). ArcelorMittal has CR production lines in France at Florange ([…]), and at Piombino in Italy ([…]). Moreover, ArcelorMittal has distribution facilities within proximity to Ilva's facilities in North-West Italy, in the area between Milan and Venice and on the French coast around Nice.

(802)              The assessment of the relevant geographic market (Section 7.2) suggests that there are elements of geographic differentiation between Southern Europe and the rest of the EEA. Such elements of differentiation also apply for CR. ArcelorMittal and Ilva are both present in Southern Europe, where they operate the two largest CR production facilities. Their main competitors' activities in CR are focussed on Northern Europe. Only Marcegaglia has a comparable, though lower, sales share in CR in Southern Europe. The other competitors do not exceed [5-10]%.

(803)              This is in line with feedback from market participants during the preliminary investigation, who indicated that ArcelorMittal and Ilva closely compete in the production and supply of CR products in Italy and surrounding countries. (583) A customer for instance points out that ‘ILVA and AM are the only European mills supplying regularly the Southern Europe, with a big range of products (HR, HDG, EG, CRC, HRP)'. (584) This has been confirmed in the Phase II market investigation, where a large majority of customer respondents (67%) indicated that they consider the range of CR products produced by Ilva to be comparable to that produced by ArcelorMittal. (585)

9.5.2.4.  Other Southern European producers are unlikely to exert sufficient competitive pressure on the parties

(804)              As regards other competitors in Italy and Southern Europe, neither Arvedi nor Marcegaglia are likely to exert sufficient competitive pressure on the parties.

(805)              It appears that Arvedi, (586)which is located in Italy, is not regarded by market participants as a close competitor to Ilva and ArcelorMittal. For the reasons  discussed in Section 9.4.2.4 on HR steel, Arvedi is not considered a viable supplier for larger volumes, also for CR.

(806)              As regards Marcegaglia, the other large steelmaker with facilities in Italy, it is not regarded as a close competitor to the Parties in view of the fact that it is a re-roller which depends on external sourcing of HR coils to manufacture secondary, whereas there are indications that integrated suppliers can guarantee a different level  of service and supply, even during times of shortage on the merchant market for steel products. Furthermore, as discussed in more detail in Section 9.7.4.2 on HDG, the Commission considers that there are notable structural and economic links between the Parties and Marcegaglia. The Transaction would further strengthen those links [commercial information]. Therefore, the Commission considers that it is likely that Marcegaglia’s ability and incentive to compete against the merged entity on the downstream market for CR would be weakened as a result of the Transaction.

(807)              In this regard it is noted that in fact both Marcegaglia and Arvedi have limited  market shares of CR in the South of Europe. As observed in Table 24: Marcegaglia has just [10-20]% of CR sales in Southern Europe and Arvedi a minor [0-5]%. This compares with the Parties' much greater combined sales share of [40-50]%. (587)

9.5.2.5. ArcelorMittal’s and Ilva’s customers overlap in the areas where the Parties compete against each other

(808)              The Commission notes that closeness of competition appears to be confirmed by an analysis of the customer overlaps between the Parties. Such closeness can be best appreciated when considering these overlaps from the perspective of Ilva's sales, as Ilva is a smaller supplier with a more limited geographic footprint and by definition the share of common sales compared to ArcelorMittal's sales will be limited.

(809)               Based on data submitted by the Notifying Party, as shown in the tables below in this recital, […]% of Ilva's CR volume is supplied to common customer locations, whilst […]% of ArcelorMittal's CR volume is supplied to these locations. Furthermore, […]% of Ilva's CR volume was supplied to common customer groups, whilst […]% of ArcelorMittal's CR volume was supplied to these same groups. Finally, […]% of CR customer groups common to both Parties were supplied by Ilva, whilst […]% of these common groups were supplied by ArcelorMittal.

[Parties' customer data]

Source: […]

(810)               In  respect  of  the  Notifying  Party's  analysis,  the  considerations  made  in   Section 9.4.2.5 in relation to HR also apply to CR.

(a)   The Notifying Party's analysis underestimates the competitive pressure of the Parties on each other. Even if one of the Parties does not conclude any sale  with a specific customer location or customer group it may still exerts a competitive constraint on the other Party to the extent that it is considered as a potential supplier. This effect is not captured by the Notifying Party's analysis.

(b)   The Commission considers that in light of ArcelorMittal's wider network in   the EEA (discussed in Section 5.2.1) it is not appropriate to base a customer overlap analysis on ArcelorMittal's sales in the entire EEA. Indeed, Ilva's sales are geographically concentrated in Italy and other neighbouring countries, whilst ArcelorMittal's sales are more dispersed all over the EEA, due to its much wider network of plants. Therefore, only the customer overlap analysis based on Ilva's footprint can provide meaningful insights for the assessment of closeness of competition between the Parties.

(811)              Taking into account the limits of the Notifying Party's analysis in capturing the competitive constraints of the Parties on each other, the Commission considers that the overlap in terms of customers is indicative of close competition between the Parties at least in the geographic area of influence of Ilva, in particular in Italy.

(812)              On the basis of the available evidence, the Commission therefore finds that ArcelorMittal and Ilva closely compete in the production and supply of CR.

9.5.3. Ilva is an important competitive force

9.5.3.1. The Notifying Party's views

(813)              The Notifying Party further argued that Ilva is not a maverick. In particular it submitted that whilst the Horizontal Merger Guidelines associate the concept of an ‘important competitive force’ with aggressive new entrants or innovators, neither applies to Ilva, whose business has demonstrably failed. It argued that Ilva's market share has  declined  in  recent  years,  from  [10-20]%  in 2012  to  around  [5-10]%  in 2016. Furthermore Ilva is capacity-constrained. According to the Notifying Party, Ilva is running at […]% blast furnace capacity utilization […]. Ilva's prices are close to import levels and reflect its deteriorating service quality. In these circumstances, the Notifying Party submitted that there is no basis to assume that Ilva exercises significant competitive constraint.

(814)              In its Reply to the SO, the Notifying Party reiterated its position that the SO wrongly characterises Ilva as an important competitive force. These arguments are summarised in Section 9.4.3.1 on HR and they are also pertinent to CR.

9.5.3.2. The Commission's assessment

(815)              First, similarly to what has been discussed in Section 9.4.3.3 on HR, also with regard to CR steel, Ilva has significant capacity which can impact EEA markets. Ilva has a total capacity for CR of […], which accounts for about […]% of total EEA CR rolling capacity.

(816)              Second, even if Ilva's primary steel (slab) production is limited due to compliance with environmental regulation, it remains amongst the largest EEA producers even at current production levels. It fulfils […]% of EEA CR demand and […]% of Southern European CR demand in 2016. In terms of CR sales, it is the fourth largest steel EEA steel supplier.

(817)              Third, even if Ilva is running at […]% blast furnace capacity utilisation […], this is a temporary situation, given the environmental restrictions which Ilva is currently subject to. This will be resolved when environmental remediation measures have been put into place. If anything, this demonstrates that Ilva's current market shares are likely understated and that they will likely increase in the future.

(818)              Fourth, as discussed for HR steel at Section 9.4.3.3 above,  Ilva's  ability to impact the EEA markets for flat carbon steel was best demonstrated when, due to environmental compliance issues, it had to significantly scale down the production of flat steel. [Internal document] (588), […].

(819)              Fifth, the Notifying Party has argued that Ilva's market share has declined in recent years, from [10-20]% in 2012 to around [5-10]% in 2016. However, as evidenced by Table 38 and Table 40, Ilva's consumption shares for both the EEA and Southern Europe have remained relatively stable between 2014 and 2016, demonstrating a slight increase in 2016 over 2015.

(820)              Sixth, as regards the Notifying Party's comments as to whether Ilva may be considered to be a maverick or not, Ilva has a history of supplying high volumes at low prices. This is reflected in the feedback received by the Commission from both customers and competitors in the course of the market investigation. (589)

(821)              Finally, as noted by the Notifying Party, the Horizontal Merger Guildelines refer, among the factors that influence whether non-coordinated effects are likely to result from a merger, inter alia to situations where the merger eliminates an important competitive force. This includes situations where '[f]or instance, a firm may be a recent entrant that is expected to exert significant competitive pressure in the future on the other firms in the market' or where two important innovators merge. However, these are only examples of situations where a merger would eliminate an important competitive force. By definition, the Horizontal Merger Guidelines do not exhaustively list every situation which could be considered to fall into this category. There can also be other situations where a firm has more of an influence on the competitive process than its market shares would suggest.

(822)              In conclusion, in view of the evidence presented in this Section, the Commission finds that Ilva is an important competitive force, which exerts more influence on ArcelorMittal on EEA markets for the production and supply of CR products than its market share suggests.

9.5.4.  EEA suppliers have no ability or incentive to constrain the merged entity

(823)               The Commission considers that EEA suppliers would not have the ability or the incentives to constrain a price increase for CR by the merged entity post-Transaction. This is because the considerations described in Section 9.4.4 regarding HR apply to CR as well. In particular:

(a)              EEA suppliers are mostly present in Northern Europe; the only competitor with a significant market position in Southern Europe is Marcegaglia;

(b)             Spare capacity is concentrated mostly in Northern Europe, (590)

(c)              Competitors are likely not to have the incentives to increase volumes in reaction to a price increase of the merged entity (see recital (617)), as shown by the competitors' behaviour in recent years of relatively high margins (see recitals (618) and (619));

(d)             As prevailing prices in Northern Europe are higher than in Southern Europe, competitors located in Northern Europe have fewer incentives to divert volumes to Southern Europe, even in the event of a price increase in Southern Europe.

9.5.5. Imports are not a sufficient constraint to offset the non-coordinated effects of the Transaction

9.5.5.1. The Notifying Party's views

(824)              First, the Notifying Party submitted that as with HR, commodity CR supplies are constrained by imports, which have grown significantly in Europe, having a  [20-30]% share of the EEA market in 2016.

(825)              Second, the Notifying Party submitted that imports will continue to exert a constraint with anti-dumping duties in place because of the substitution effect offsetting the decrease in imports from countries affected by anti-dumping measures by unaffected countries such as  Turkey,  India  and  South  Korea. (591)  Average  monthly  imports  in January to August 2017 were identical to the same period in 2015. Following a dip in 2016, the import share rebounded in 2017 to above [20-30]%. (592)

(826)              Third, the Notifying Party argued that EEA steel prices are a function of the global supply/demand balance. According to the Notifying Party, anti-dumping duties have not undermined this constraint: imports have been redirected from affected to non- affected exporting countries.

(827)               Fourth, the Notifying Party argued that imports are, and will remain, the greatest constraint on Ilva. It submitted that a large majority of imports land in Southern Europe and in Italy in particular. Large volumes of global divertible capacity can be used to profitably supply the EEA, under a variety of price levels and including the impact of anti-dumping duties. (593)

(828)               Fifth, the Notifying Party noted (594) that the Commission's preliminary market investigation has in general confirmed that a very large number of EEA customers rely on imports as a source of supply. In this regard, according to the Notifying   Party, an analysis of EEA customers' share of CR steel purchases from non-EEA producers shows that:

(a)              Customers source anywhere between 0–100% of their requirements from imports;

(b)             Even small customers can source a significant share of their needs from non-EEA producers;

(c)              The share of imports can sharply change over time. Some customers have decreased purchases from imports between 2015 and 2016.

(829)               The Notifying Party therefore concluded that i) customers can credibly threaten to switch further volumes to imports and ii) imports have not reached a ‘natural limit’.

Figure 96 - ArcelorMittal submission on CR import levels

[…]

(830)              Sixth, with regard to competitive constraint, the Notifying Party argued that a comparison of ArcelorMittal's CR prices to import prices suggests that ArcelorMittal's prices remain highly correlated with import prices. Any weakening of the competitive constraint exerted by imports could be expected to be reflected in an increase in ArcelorMittal's premium over imports. However, ArcelorMittal's spread over import has not increased in recent times. This would, according to the Notifying Party, indicate that there is no evidence of a weakening of the constraint by imports.

9.5.5.2. The Commission's assessment

(831)              For the reasons set out in this Section, the Commission considers that, contrary to the Notifying Party’s submission, imports are not a sufficient constraint to offset the non-coordinated effects of the Transaction. In this regard, the considerations elaborated in relation to imports at Section 9.4.5., also apply to CR.

(832)              The results of the Commission's investigation have in general confirmed that a number of CR customers in the EEA use importers as a source of supply, particularly for commodity grades. However, the Phase II market investigation has confirmed  that market participants do not consider imports to be fully substitutable with domestic supplies of CR. The biggest issue in relation to non-EEA imports of CR are the much longer lead times, which 85% of respondents indicated to be worse as compared to EEA based deliveries. Half of the respondents are concerned about the reliability of deliveries from outside the EEA and 36% of respondents are concerned about availability. One third of industrial customer respondents have indicated that such issues would actually prevent them from importing CR commodity products from outside the EEA. (595)

(833)              In line with the feedback for HR, whilst imports from China have come to a  standstill, the indications are that imports from other countries, replacing Chinese volumes, are coming into the EEA at a higher price than a year ago. Over one third  of industrial customer respondents who responded to the investigation (35%) have experienced price increases for non-EEA imports of CR in 2017, as against only 8% who did not. (596) The reasons respondents give for this include: the fact that there have been anti-dumping duties imposed on products from a number of countries and the perception that the price of steel has also increased outside the EEA. It would therefore appear, inter alia, that the removal of the market perturbation caused by dumping from certain non-EEA countries had an effect on the price at which CR products come into the EEA. (597)

(834)              A minority (15%) of respondents to the Commission's investigation indicate that they have switched their imports of CR away from countries affected by anti-dumping measures (Russia and China) towards other countries. (598)

(835)              Considering CR steel, a large number of customers taking a position have indicated that a small but significant increase in the prices of EEA-suppliers for CR would either not prompt them to switch to imported CR at all or to only switch part of their sourcing. When asked how they would react to a price increase by ArcelorMittal, almost half of respondents indicated that in this case they would source from another producer within the EEA, rather than from non-EEA producers (21%). (599)

(836)              The Commission considers that if the merged entity increased prices of CR post Transaction, imports would not be a sufficiently strong constraint to offset and defeat such price increase. As discussed in Section 9.4.5 for HR, this is due to a number of reasons, which are equally applicable to CR. Firstly, following the removal of the perturbation caused by dumping and the restoration of fair trading conditions, there appears to be an upward pressure on the price of imports. Secondly, given the limit  as to how much EEA customers are willing and able to source from imports, whilst imported CR may exert some pressure on the prices post-Transaction, the merged entity would retain a guaranteed share of supply, which would give it the ability to increase prices, at least on that part of demand where their supply cannot be replaced. Imports can only contest a part of the demand of EEA customers.

(837)               The ability of non-EEA suppliers to reliably and continuously serve EEA markets is not only limited by their focus on domestic sales, but also by the characteristics of  the demand by EEA customers. The Commission's investigation indicates that only a limited portion of CR demand in the EEA could be challenged by imports.

(838)               On the basis of ArcelorMittal's [commercial information]. (839) Looking at Southern Europe, [commercial information].

(840)    Furthermore, it is notable that a large majority of industrial customers who responded to the market investigation have indicated that the maximum percentage of CR commodity products that they could realistically source from outside the EEA is between 0–10%. (600) In practice, the majority of industrial customers who responded to the Commission's market investigation only source between 0–5% of their annual requirements in CR from outside the EEA. If combined with the 14% of industrial customers who source  between 5–10% of their annual CR requirements outside     the EEA, that means that 75% of the respondents source a maximum of 10% of their annual CR requirements from outside the EEA. (601) This feedback is not in line with  the Notifying Party's argument that customers source anywhere between 0–100% of their requirements from imports.

9.5.5.3.  Trade defence measures

(841)    The Commission considers that the assessment of the ability of imports to exert a constraint on the merged entity must be made in light of the applicable trade defence measures.

(842)    In particular, the recently adopted trade measures against major steel exporting countries in order to restore fair market conditions may result in a lower ability of imports to exert competitive pressure on EEA-based CR producers in the future as compared to the competitive pressure exerted by imports when dumping happened. Market shares of imports until 2016, affected by dumping, may therefore overstate the imports' ability to constrain the merged entity in the near future.

(843)    On the basis of the data available to the Commission, it appears that, following the provisional anti-dumping duties adopted by the European Commission against Chinese suppliers of CR in February 2016, imports of CR into the EEA from China have substantially decreased.

(844)    ArcelorMittal submitted that these amounts can be replaced by imports from other non-EEA countries. According to ArcelorMittal it would appear that in 2017 imports of CR reached back to 2015 levels. The Figure 97 shows the CR import levels on a monthly basis in the EEA, based on a submission by ArcelorMittal.

Figure 97 – ArcelorMittal submission on CR import levels

[…]

(845)    Based upon its findings, and on the most recent available data, the Commission observes that CR imports have increased from 2016 to 2017 but are still below the record 2015 values. There are therefore indications that despite a substitution with volumes from other countries, imports have not been able to reach the 2015 peak levels.

(846)    Further the Commission observes that imports of CR in the second part of 2017 have reduced as compared to the first part of 2017 by a factor of 17%. On the contrary in year 2016 imports of CR grew by 20% between the first and the second half of the year (Figure 99). Therefore the most recent trends indicate an attenuation of the inflow of CR imports into the EEA.

(847)    Moreover, as discussed in the Section on HR, imports have come at higher prices, as price levels for flat carbon steels in the EEA have increased significantly.

(848)    Feedback received during the Commission's market investigation indicates that to the extent the imports from the countries affected by the anti-dumping duties have been replaced by imports from other non-EEA countries, the CR imported from those non- affected countries, which did not engage in dumping practices, has been sold to EEA customers at prices significantly higher compared to the prices that were offered by suppliers from countries affected by the anti-dumping measures. (602) Any substitution of imports from other non-EEA countries has thus taken place at a higher price level, indicating that the constraint exerted by alternative sources of imports on pricing has become weaker.

(849)    Similarly, a clear majority of customers confirmed the increase in the prices of non-EEA imports. Some customers have further explained that the increase in price was at least partially due to the trade defence measures. (603)

9.5.5.4. Importers generally pursue an opportunistic strategy

(850)    In line with what has been discussed in Section 9.4.5.5 for HR, there are indications that non-EEA CR suppliers follow an opportunistic strategy, whereby their supplies to the EEA are a function of fluctuations in price and considerations on prevailing conditions of demand in their own domestic markets.

(851)    While the Notifying Party submitted that there is large capacity also of CR at the global level that can be diverted and imported into the EEA, the Notifying Party was not able to identify the individual exporters, with the exception of a few.

(852)    Moreover, the Commission notes that there have been fluctuations in import levels of CR over recent  years, for example between 2009–2010 and in 2012, as depicted  in Figure 98.

(853)    As such, the historical data suggests that the CR import flows into the EEA follow cyclical patterns. This further indicates the opportunistic nature of imports and the fact non-EEA imports might not exert a constant competitive pressure on EEA-based domestic steel producers. Such opportunistic behaviour is also reflected in the commercial conditions for steel purchases. Market participants have confirmed that the sourcing of CR from importers is mostly on a spot basis with limited mid or long- term commercial agreements. (604) Almost half of industrial customers who responded to the second market test procure non-EEA imports on the basis of spot purchases. This compares with less than one fifth who procure from outside the EEA on the basis of midterm agreements and 6% on the basis of a long term agreement.

(854)    Such opportunistic behaviour means that the non-EEA producers are a variable source of supply that does not provide the same reliability and availability of CR products as EEA-based producers.

(855)    An analysis of spreads in various markets can offer an overview of the market conditions at different potential markets for non-EEA producers. Figure 100 illustrates the price spread over raw material prices for CR in various countries, notably the US, China, Turkey and Europe (South and North).

Figure 100 – CR spread over raw materials

[…]

(856)    As can be observed from the graph in Figure 100 the Turkish domestic spread  is often above the spread in Southern Europe, although this appears to have diverged below the Southern European spread between 2016 and 2017. Feedback from the market investigation suggests that capacity in Turkey is mainly used for the domestic Turkish market. One Turkish producer has explained that 'As a general consideration, in this moment the EEA market is not particularly attractive and profitable for Turkish steel operators which prefer to export their products in other countries.' That producer, for example, does not export more than 25% of its total production, as it concentrates on the domestic Turkish market. The volumes it sells into the EEA are very limited. (605)

(857)    The only spread consistently below Southern Europe spreads is the Chinese domestic price spread over raw materials. However, as discussed in relation to imports for HR in Section 9.4.5.7, since the implementation of EU anti-dumping measures against China for CR steel products in 2016, the indications are that imports from that region for CR steel have effectively ceased.

9.5.5.5. Compared to domestic producers, imports are a less reliable source of supply for EEA customers

(858)    All factors considered, it is clear that EEA based customers have in any case a preference to procure CR steel from within the EEA and use imports, for the most part, as a secondary source of supply. It would appear that whilst imports of CR from non-EEA countries would act to some extent as a constraint on the merged entity in the production and supply of CR post-Transaction, such constraint would be limited in nature. Therefore, in the Commission's view, imports would not constrain the merged entity sufficiently as to defeat its incentives to decrease prices post- Transaction.

9.5.6. The competitive constraint by northern suppliers and imports in Southern Europe is limited

(859)    Furthermore, contrary to ArcelorMittal's claim that northern suppliers can and do supply in Southern Europe, the presence of such northern suppliers is limited.  Indeed, this is also evidenced by these suppliers' negligible market shares for CR sales in Southern Europe, as observed in Table 39, where individual market shares for northern suppliers range between just […]% and […]%. (606)

(860)    The Commission notes that in 2016 imports constituted [20-30]% of the sales share  of CR in Southern Europe. However, as discussed in more detail and for the reasons given at Section 9.5.5, whilst imports of CR from outside the EEA would to some extent act as a constraint on the merged entity in the production and supply of CR post-Transaction, this constraint would be limited.

9.5.7. Customers are concerned about the effects of the Transaction

(861)    In terms of impact, the results of the market investigation confirm that there are concerns about alternative sources of supply of CR post-transaction. A clear majority of those customers who expressed an opinion indicated that should the Transaction  go ahead, they expect to  see  an  increase  in  the  price  of  CR  in  general  across the EEA. (607) Several respondents indicated they would expect to see a stronger impact in the South of Europe, more specifically in Italy. (608) Moreover, several industrial customer respondents who expressed an opinion believe the deal would reduce the current choice of CR supply available to the market in that it would be more difficult for the customers to find alternative CR suppliers. (609) Several customers further indicated that, at least as concerns ArcelorMittal, post-Transaction they would expect prices for CR to be the same in Southern Europe as the North, given that  the company implements a centralised pricing policy. (610) In other words, they would expect to see a price increase in Southern Europe. Finally, even competitors of the Parties expect the Transaction to have an impact on the price of CR in the EEA in that the prices of CR will increase. (611)

(862)    Customers that responded to the Commission's market investigation highlighted the fact that the Transaction would strengthen ArcelorMittal's current market leading position in the EEA and would lead to less competition in an already concentrated market, with consequences for both sourcing and pricing. (612) For example, one industrial customer has indicated that the '[a]cquisition of Ilva by ArcelorMittal reduces options within the European market for sourcing steel products. With recent anti-dumping duties applied by the European Commission on all main steel producers around the globe, European steel producers selling power has increased greatly. We are already facing the consequences of the above mention situations as steel producers are pushing for a price increase for next year volume of 10%+ whereas raw material cost increase should limit the price increase to 4% maximum. This supply/demand situation (with competition from non-European producers cut due to anti-dumping duties) will only be more and more difficult for customers due to the European market being further and further consolidated with the recent announcement of Tata & Thyssenkrupp to create a JV. '(613) This is reflected in comments made by numerous other industrial customers

(863)    A number of customers of CR steel have also indicated that they are concerned that the Transaction would lead to an increase in prices across the EEA, even beyond commodity products. As an illustration, one customer indicated that it 'expects the transaction to have potential impact on its business because with the Ilva acquisition ArcelorMittal – due to its size with Ilva - could become a total ruler of the steel market in the EU. More generally, [ ] does not want to judge the impact, yet  confirms that while Ilva is not yet a major Auto player, due to its size, it could still influence the prices in EU for Auto steel'. (614)

(864)    There are also indications from the market investigation that some customers fear the Transaction would lead to an increase in prices for CR products more specifically in Southern Europe. (615) For example, when asked about what they thought the impact of the Transaction would be, a customer indicated that '[i]n terms of market dynamics,  [ ] notes that the European steel industry is divided into North and South, with different price levels. Historically, Southern Europe has been more competitive price-wise; Ilva is the most important player in this region. […] transaction could have an impact in terms of driving up prices on the Southern market. Ilva has been known as an aggressive player on prices and therefore it remains to be seen whether this will change post-transaction. (616)

9.5.8.  Conclusion on horizontal non-coordinated effects in the market for the production and supply of CR flat carbon steel products

(865)    With regard to the market for the production and supply of CR flat carbon steel products in the EEA, the Commission finds that the Transaction would strengthen ArcelorMittal’s market position, both in terms of sales and, structurally, in terms of CR capacity within the EEA.

(866)    The Transaction would eliminate the competitive interaction between ArcelorMittal and Ilva, who compete very closely in Southern Europe.

(867)    The Commission also finds that the constraint exerted by Ilva on ArcelorMittal is significant, as Ilva has acted as an important competitive force in view of its size and strategy.

(868)    Based on evidence gathered during the market investigation, the Commission also finds that the remaining EEA suppliers would not have the incentive and ability to constrain the merged entity. Similarly, imports would not be likely to sufficiently constrain the merged entity so as to avoid the likely effects of the Transaction on customers.

(869)    In conclusion, based on the information provided by the Notifying Party and on the results of the market investigation, the Commission considers that the Transaction is likely to significantly impede effective competition through the removal of an important competitive constraint in the market for the production and supply of CR flat carbon steel products in the EEA as a result of horizontal non-coordinated effects.

9.6. Production and supply of galvanised flat carbon steel products (GS): horizontal non-coordinated effects

(870)    The Parties’ activities overlap in the production and supply of GS within the EEA as well as in Southern Europe, including sales in Italy. An overlap similarly arises with regard to both of the sub-segments of GS, that is HDG and EG.

(871)    For the reasons set out in Section 6.5, the Commission finds that that there is at least a serious possibility that HDG and EG constitute distinct product markets. Therefore, in this Decision, the Commission will assess sub-segments HDG and EG separately.

(872)    Concerning the overall supply of all GS – that is both HDG and EG – the Commission observes that HDG accounts for over […]% of the Parties’ total  supplies of GS in the EEA. Based on information provided by the Notifying Party, the overall split between HDG and EG in the supply of GS within the EEA is similar to the split in the Parties’ own supplies. The structure of supply of GS and the shares of HDG and EG is shown in Table 43.

(873)    Therefore, and in light of the evidence available to it, the Commission considers that even if the product market suggested by the Notifying Party – that is an overall market for all GS – was considered, the findings of the market investigation and the outcome of the assessment would be similar to the findings explained and the outcomes reached in this Decision for HDG in Section 9.7.

9.7. Production and supply of hot-dip-galvanised flat carbon steel products (HDG): horizontal non-coordinated effects

9.7.1. The merged entity will be the largest HDG producer, with increased market power,  in the EEA

(874)    Prior to the Transaction, ArcelorMittal is the market leader, both in terms of sales market share and capacity share in the EEA as well as in Southern Europe, including in Italy.

(875)    At the EEA-level, ArcelorMittal had a sales market share of [30-40]% in 2016 while the Parties achieved a combined market share of [30-40]%. The largest competitor, ThyssenKrupp, remained at [10-20]% sales market share. No other EEA-producer achieved 10%  share  at  the  EEA  level  in  2016,  and  imports  remained  in  total  at [10-20]%.

(876)    In Southern Europe, ArcelorMittal had a sales market share of [30-40]% in 2016 while the Parties achieved a combined market share of [40-50]%, Ilva adding [10-20] percentage points to ArcelorMittal’s market share.  No  other  EEA-producer achieved 10% share in Southern Europe in 2016. The largest competitor, re-roller Marcegaglia, had [5-10]% sales market share in 2016. Imports remained at [10-20]% of the total sales in Southern Europe.

(877)    In Italy, the combined sales market shares of the Parties was [30-40]% with ArcelorMittal's and Ilva's market shares respectively at [10-20]% and [10-20]%. Arvedi would be the third player following closely the merging parties with a market share of [10-20]%.

(878)    The Parties’ and their competitors’ sales market shares in the EEA and in Southern Europe are shown in Table 44, which reproduce information provided in Table 21, Table 24 and Table 25.

(879)    In terms of capacity, ArcelorMittal’s position in HDG in the EEA in 2016 was even stronger than in terms of sales: ArcelorMittal had a capacity share of [40-50]%– [30-40]% (Scenario 1 – Scenario 2) (619) in the EEA in 2016. The Parties combined had in total had [40-50]% of the capacity in the EEA in 2016 with an increment of [5-10] represented by Ilva’s share.

(880)    For completeness, the Commission also provides shares for GS considered as  a whole (HDG + EG). The data are similar as for HDG: ArcelorMittal had a capacity share of [40-50]% – [30-40]% (Scenario 1 – Scenario 2) in the EEA in 2016 while the Parties in total [40-50]% of the capacity, Ilva adding [5-10] percentage points to ArcelorMittal’s share.

(881)    The Parties' capacity shares in the EEA are shown in Table 45.

(882)    Unlike HR and CR, the majority of HDG produced in the EEA is sold to third parties instead of being used captively by the steel producers. Therefore, the Commission does not consider consumption shares as instructive in assessing the HDG. In any event, given that the majority of HDG produced is sold to third parties, the consumption shares tend to be largely similar to the sales market shares.

(883)    The Notifying Party submitted that despite the Parties’ high combined market shares, the market share increment is small and not indicative of competition concerns. Moreover, in the view of the Notifying Party, there are numerous alternative suppliers present, including imports.

(884)    Nonetheless, the Commission observes that ArcelorMittal has by far the strongest market position both in the EEA and in Southern Europe (including in Italy where ArcelorMittal is only second to Ilva) already before the Transaction. Ilva is a significant producer, being the 5th largest EEA-producer at the EEA level and the largest EEA competitor to ArcelorMittal in Southern Europe in terms of sales shares. The findings for capacity shares are similar.

(885)    The market share increment brought by the Transaction is significant. In Southern Europe, the Transaction even constitutes a merger between the two largest producers and suppliers. The Transaction will thus increase ArcelorMittal’s market position further with all remaining competitors being significantly smaller in terms of sales market share and capacity share. Therefore, the Commission considers the merged entity is likely to have increased market power in the EEA and even more so in Southern Europe.

9.7.2. ArcelorMittal and Ilva are close competitors in HDG

9.7.2.1. The Notifying Party’s views

(886)    The Notifying Party submitted that the Parties are not particularly close competitors in HDG. According to ArcelorMittal, the Parties’ product offering is complementary: ArcelorMittal  submitted  that  its  own  focus  is  on  high-end  value-added products whereas Ilva has historically manufactured commodity products. ArcelorMittal acknowledged that Ilva’s production line in Novi Ligure is of higher quality and has an extended range of products for the automotive industry; however, according to ArcelorMittal, the production volumes in Novi Ligure are very small. ArcelorMittal also submitted that the customer overlap between the Parties is limited, that customers have buyer power and that they can easily switch suppliers.

(887)    Nonetheless, for the reasons set out in this Section, the Commission considers that the Parties are close competitors in the market for HDG.

9.7.2.2. The Parties are both integrated steel manufacturers

(888)    The Commission recalls that both of the Parties are integrated steel producers active in the whole steel value chain upstream of HDG. In practice, both ArcelorMittal and Ilva produce the steel substrate used in their production of HDG (that is HR and CR). This makes them close competitors as an integrated producer can better reply to the competitive challenge of another integrated producer than what a non-integrated producer (re-roller) can.

(889)    In the market investigation, the majority of customers responding indicated that non- integrated suppliers are not as competitive in meeting the customers’ needs for HDG as integrated suppliers are. A number of customers indicated that parameters such as stable quality can be better ensured by a vertically integrated supplier and that such factors have importance. An industrial customer explains: ‘From our Point [sic] of view it is important that a EEA supplier for us is vertically integrated. Reasons consisting of Substrate, availability of grades and stable Quality’ while another industrial customer notes that ‘[i]t is hard to meet stricter needs when you do not have your own primary steel production. We have bad experience  with  Marcegaglia.’ These views have been shared by a number of distributors as well; a distributor observes: ‘[a] re-roller would have to buy HR coils. Therefore it will be dependent on other EU mills or Import.  Both  prices  and  quality  could  suffer  from it. (621)

(890)    Furthermore, a non-integrated supplier depends on the availability and pricing of HR (or CR to which HR is the main input) on the merchant market, whether produced in the EEA or imported. In this respect, the Commission recalls that HR has been recently subject to a number of trade defence measures, as explained in Section 5.3.3 and 9.4.5. To this effect, a customer also noted that ‘[p]rovided there are no trade barriers, the answer is YES [non-integrated producer can be competitive]. One can always buy competitively priced steel on the world market. With trade barriers in place, re-rollers have a more difficult position’. (622) In line with this, an HDG supplier that currently produces HDG within the EEA but uses imported HR input has submitted that it ‘[i]n light of the antidumping duties, [name of the supplier] aims to modify its commercial strategy and to sell its galvanised products outside the EEA to ensure company’s profitability.’ (623)

9.7.2.3.  Ilva’s product offering competes directly with that of ArcelorMittal’s

(891)    The Notifying Party submitted that Ilva is primarily a commodity producer,  including with respect to its HDG production. This view was shared in the market investigation by the majority of customers although a number of customers considered that Ilva produces high-end products as well. (624)

(892)    Nonetheless, the Commission observes that all HDG producers typically produce some commodity products even if they preferred to concentrate on high-end  products. This is in line with the considerations in Section 6.10 and, to that effect, also ArcelorMittal acknowledges that a steel manufacturer cannot in practice only produce high-end products but that periods of ‘relaxation’ are required. According to ArcelorMittal, this extends to HDG production where, for instance, specific conditions in the zinc bath might be required for high-end products. In line with these considerations, ArcelorMittal explains with respect to its own production that, as an example, automotive exposed parts (that is high-end products) account for at most […] of the output of a given galvanising line. (625)

(893)    Concerning ArcelorMittal’s actual production, and based on information provided by ArcelorMittal, slightly over […]% of its GS production – […] – is commodity products, a […] as shown in Figure 101.

Figure 101 – ArcelorMittal's GS production

[…]

Source: Commission's computation on the basis of the Parties data and Parties' own classification of Commodity/Specialty product

(894)    Therefore, unlike the suggestion made by the Notifying Party, both of the Parties produce significant volumes of commodity products and their product portfolios are not exclusively complementary in the sense that one of them would only produce commodities and the other high-end products.

(895)    Furthermore, the Commission considers that, even if Ilva primarily produces commodity HDG, its importance in HDG extends beyond commodity products: For the reasons explained in Section 5.4, commodity production can indirectly act as a constraint for the pricing of high-end products. In other words, even if Ilva only produced commodities, quad non (see Section 9.7.2.4) it could exert a level of competitive constraint on high-end producers, including the non-commodity HDG business of ArcelorMittal.

9.7.2.4. Ilva is capable of producing quality HDG

(896)    Ilva appears to be able to produce at least some products to demanding end applications, such as car manufacturing. The Commission recalls that, unlike many  of its other production lines, Ilva’s GS lines, and particularly HDG lines, are relatively new. In particular, Ilva has built new HDG production lines in Novi Ligure and Genova between 2010 and 2012, commencing production on one of them as late as in 2016. [Internal document]. (626)  […] (627) […].

Figure 102 – Extract from ArcelorMittal internal document (628)

[…]

(897)    [Internal document]. (629)

Figure 103 – Extract from ArcelorMittal internal document

[…]

(898)    The results of the market investigation indicate that the quality of Ilva’s HDG products support its ability to be an important competitive force and a close competitor to ArcelorMittal. In fact, the majority of both industrial and distributor customers responding suggested that the quality of Ilva’s HDG products is at least comparable to that of other HDG suppliers, such as ArcelorMittal. (630) Customers describe Ilva’s HDG quality, for instance, as ‘very good’, ‘good’ and  ‘satisfactory’. (631) A distributor customer further explains that ‘[t]he quality of HDG produced by the Novi plant is very good; has lately become a bit worse because of the quality of the HRC material - - but the HDG production facility of Novi is  modern - - the HDG produced in other Ilva facilities are less “beautiful”.’ (632) A car manufacturer, despite noting that Ilva’s performance is somewhat inferior to other producers, explains that Ilva has no major quality problems and that Ilva’s quality ‘is in line with our expectations’. (633)

9.7.2.5.  Parties’ HDG geographic footprints largely overlap

(899)    The Parties are two of the few major integrated flat carbon steel manufacturers that have HDG production sites in Southern Europe, including in Italy, which is the   area where the competitive interaction between the Parties is most important: Ilva has production sites in Italy while ArcelorMittal has them in Italy and Spain. Out of the other integrated EEA flat carbon steel manufacturers only Arvedi has plants in Italy and ThyssenKrupp one satellite plant in Spain. In addition, the Brazilian integrated supplier Lusosider (CSN) has one satellite plant in Portugal. The rest of the suppliers in Southern Europe are non-integrated producers including, first and foremost Marcegaglia in Italy.

(900)    Furthermore, [internal document]. (634)

9.7.2.6. ArcelorMittal’s and Ilva’s customers overlap in the areas where the Parties compete against each other

(901)    In light of the available evidence, and contrary to the Notifying Party’s submissions, it appears that asignificant amount of the Parties’ (actual and potential) customers overlap. In particular, it seems that the areas where Ilva is present and supplies its products are characterised by a significant customer overlap and customer switching between the Parties, which indicates an intense competitive interaction between ArcelorMittal and Ilva at least in some significant parts of the EEA and particularly  in Southern Europe region, including in Italy.

(902)    The Notifying Party has submitted an analysis on customer overlap concerning the overall GS products. The Commission considers that, given the fact that the clear majority of GS sold is HDG (see Table 43) the analysis can be used as a proxy for HDG as well.

(903)    In its analysis, the Notifying Party submitted that for GS, considering ArcelorMittal's customer base, only […]% of the sales are related to common customer locations ([…]% considering customer groups). (635) On the other hand, considering the customer base of Ilva, […]% of Ilva's sales  are  related  to  common  customer  location  ([…]% considering customer groups/companies). (636)

(904)    The Commission considers that ArcelorMittal's customer overlap analysis underestimates the competitive pressure exerted by the Parties on each other. If anything, ArcelorMittal's analysis shows that the Parties overlap for a large proportion of Ilva's sales.

(905)    In the first place, even if one of the Parties did not conclude any sale with a specific customer location or customer group it might still exert a competitive constraint on the other Party if it is considered as a potential supplier. This effect is not captured by the Notifying Party's analysis.

(906)    In the second place, the Commission considers that it is not appropriate to base a customer overlap analysis on ArcelorMittal's sales in the entire EEA only. Indeed, Ilva's sales are geographically concentrated in the Southern Europe region, in particular in Italy and Spain, (even if it also sells volumes in other countries outside Southern Europe such as in Germany) whilst ArcelorMittal's sales are more dispersed all  over  the  EEA,  due  to  its  much  wider  network  of  plants  (discussed  in Section 5.2.1). This is also shown in the Notifying Party’s submission in that, within the EEA, […]% of its customers overlap with Ilva while in Italy the share is  […]%. (637) A customer overlap analysis based on Ilva's footprint can provide meaningful insights for the assessment of closeness of competition between the Parties. On this basis, the analysis carried out by the Notifying Party indicates that a large proportion of Ilva's GS volumes in the EEA are sold to common customers: […]% considering customer locations and […]% considering customer groups. (638)

(907)    Taking into account the limits of the Notifying Party’s analysis in capturing the competitive constraints exercised by the Parties on each other, the Commission considers that the identified overlap in terms of customers is indicative of close competition between the Parties in particular in the geographic area of influence of Ilva.

(908)    The Notifying Party has further assessed interaction between ArcelorMittal and Ilva in terms of losses and wins at customer accounts. To that effect, the Notifying Party submitted an analysis of customer accounts where ArcelorMittal has lost or gained at least […]% or at least […]% of its volumes of GS, on the one hand at the EEA-level and, on the other hand, in Southern Europe region. [Parties' submissions].

(909)    The results of the analysis submitted by ArcelorMittal on accounts where it lost volumes are largely similar to those on accounts where ArcelorMittal gained. In essence, [commercial information]. When looking more specifically at the geographic area where the competition between the Parties mostly takes place, that is Southern Europe, [commercial information] Results of the analysis are shown in Table 47.

9.7.2.7. Feedback from market participants confirms that the Parties are close competitors (910)   The results  of the market  investigation  corroborate the  finding  that  the  Parties are close competitors. In particular, a clear majority of customers taking a position in the market investigation considered that ArcelorMittal and Ilva compete closely in the price and quality of commodity HDG products. Opinions on other parameters of competition were more split, though approximately half of customers taking a position submitted that the Parties also compete closely in the range of HDG  products and in lead time. Some customers further considered that the Parties compete in high-end products though that opinion was not shared by the majority of them. (641) An Italian distributor customer explains: ‘[F]or HDG it [Ilva] can also target good high grade and surface qualities, but it lacks on service. As far as the dimensions availability, Ilva and ArcelorMittal are much the same’.

(911)    The position of competitors in the market investigation was largely similar to that of customers. A competitor explains in this respect that the Parties are ‘both active in the same regional markets with more or less same offering’ while another one notes that ‘[i]n Southern Europe AM and Ilva compete strongly in hot dip galvanized products. For differentiated products Ilva are not considered to be an important player’. (642)

9.7.3. Ilva is an important competitive force

9.7.3.1. The Notifying Party’s arguments

(912)    The Notifying Party submitted that Ilva is not an important competitive constraint. In particular, the Notifying Party suggests that Ilva’s production and supplies are ailing and it is also capacity constrained in the upstream HR.

9.7.3.2. The Commission’s assessment

(913)    For the reasons set out in this Section, the Commission finds that, contrary to the Notifying Party’s position, Ilva is an important competitive constraint.

(914)    First, Ilva has significant HDG capacity with which it can affect market conditions, particularly in Southern Europe. Ilva produces HDG at three different sites in Italy: Genova, Novi Ligure and Taranto. The installed GS production capacities of those sites are shown in Table 48.

(915)    Compared to the capacity on the market, Ilva’s capacity is not insignificant. As can be seen from Table 45, Ilva has […]% of the EEA production capacity for GS, which is mostly HDG. Only ArcelorMittal and ThyssenKrupp have significantly higher capacities than Ilva.

(916)    Second, Ilva’s production volumes have been increasing in the past years and Ilva  has invested in and brought to use new HDG capacity. The Commission further observes that while Ilva’s production of hot liquid metal and HR has been reduced due to environmental questions, there is no similar reduction in its production of HDG. In fact, while Ilva’s production of HR has decreased from […] in 2012 to […] in 2016, that is […]%, its HDG production has only experienced a significant reduction in 2013 compared 2012, resuming to and exceeding previous levels thereafter. Based on information provided by the Notifying Party, Ilva has even finalised and opened a new HDG line in 2016 – originally built in 2012 – and produced more HDG in 2016 than in 2012 before the major reduction in upstream HR production. The development of Ilva’s production during 2012–2016 is shown in Table 49.

(917)    Third, Ilva is particularly strong in Southern Europe. As can be seen in Table 44, Ilva is the second largest supplier of HDG in Southern Europe, right after ArcelorMittal. In Italy Ilva is the second largest supplier after ArcelorMittal and with higher market share than Arvedi and the re-roller Marcegaglia.

(918)    Fourth, [internal document]. (647) […]. (648)

9.7.4.EEA suppliers would not be able or have the incentives to constrain the merged  entity

9.7.4.1.  Spare capacities are limited

(919)    The Notifying Party submitted that remaining competitors will be able and are likely to constrain the merged entity. In particular, the Notifying Party submitted that a number of EEA-suppliers have adequate excess capacity to thwart price increases by the merged entity.

(920)    The Notifying Party suggested that, in the EEA, there is spare capacity of over […]  of GS and that the capacity utilization overall is around […]%. (649)

(921)    Nonetheless, the information gathered by the Commission from steel producers during the market investigation suggests significantly smaller spare capacities. Overall, the results of the market investigation suggest that the overall spare capacity of the Parties’ competitors  in  the EEA is  less  than  1  Mt  when considering readily available capacity and remains less than 1 Mt even if considering capacity that can  be taken on line in the short/medium term. The competitor’s capacity utilisation    and spare capacities relative to total capacities on an aggregate level are shown in Table 50.

(922)    [Parties' submissions]. (651)

(923)    Furthermore, it can be recalled with respect to spare capacities that Ilva sold […] of HDG in the EEA in 2016 (of which […] in Southern Europe). Therefore, the spare capacities of EEA producers other than the Parties cover […] of the production of Ilva at the EEA level, under Scenario 1, and […] of Ilva's  production  under  Scenario 2.

(924)    Furthermore, it appears that ArcelorMittal internally considers that overcapacities in HDG are diminishing. For instance […]. (652)  […] (653) […].

Figure 104 – ArcelorMittal internal document (654)

[…]

(925)    Therefore, it appears that the spare capacities available in the market are limited even at the overall EEA level. Based on ArcelorMittal’s internal documents, it furthermore appears likely that […].

(926)    Consequently, the Commission considers that EEA competitors will likely not have the ability to counter price increases by the merged entity. Nor will they have the incentive to do so, given that they would benefit from an increased price themselves in a situation of tight supplies without having the spare capacity to gain more market share if they chose to not follow  price  increases.  Moreover,  as  explained  in  recital (617) for HR, even to the extent that competitors have spare capacity, they are unlikely to have a strong incentive to expand output to such an extent as to render a price increase unprofitable for the combined entity. In that situation, while competitors may expand their output to some extent following price increases to benefit  from  increased  demand,  they  likely  have  the  incentive  to  restrain such reactions and raise their prices in turn in order to benefit from higher volume at higher prices. (655)

9.7.4.2. Marcegaglia is not a strong competitive constraint on the Parties

(927)    The Commission recalls that Marcegaglia is one of the Parties’ main competitors in terms of sales volumes and, as shown in Table 44, it is the Parties’ largest competitor in terms of HDG sales volumes in Southern Europe.

(928)    Already prior to the Transaction, Marcegaglia holds a minority stake […] at ArcelorMittal’s Bremen steelworks. The Commission understands that there are no other structural links existing between ArcelorMittal and Marcegaglia prior to the Transaction.

(929)    As explained in Section 2, Marcegaglia is a party to AM Consortium, the vehicle ArcelorMittal is using to acquire Ilva, [commercial information].

(930)    [Parties' submissions]. (656) […]. (657)

(931)    Consequently, the Commission considers that there are notable structural and economic links between the Parties and Marcegaglia. The Transaction would further strengthen those links and create a two-sided cross-ownership structure. Therefore, the Commission considers that it is likely that Marcegaglia’s ability and incentive to compete against the merged entity on the downstream markets for HDG would be weakened as a result of the Transaction.

(932)    The likelihood that the Transaction may have a negative effect on the competition between ArcelorMittal and Marcegaglia has also been suggested by a number of market participants. (658) A customer explains: ‘[B]eing part of the deal, even though with a smaller share, they will be even more influenced by Arcelormittal policy’. (659)

9.7.5.  Imports are not a sufficient constraint to offset the non-coordinated effects of the Transaction

9.7.5.1.  The Notifying Party’s views

(933)    The Notifying Party submitted that, in addition to competition from EEA producers, the merged entity would continue to face competition from non-EEA producers that import HDG into the EEA, thus preventing the merged entity from raising prices.

9.7.5.2. The Commission’s assessment

(934)    For the reasons set out in this Section, the Commission considers that, contrary to the Notifying Party’s submission, imports are not a sufficient constraint to offset the non-coordinated effects of the Transaction.

(935)    To this effect, the Commission first recalls the considerations elaborated on in detail for HR in Section 9.4.5., including among others the volatility and opportunistic nature of imports and the fact that importers are individually small. The results of the market investigation and the information available to the Commission indicate that the considerations explained in that Section apply to HDG as well.

9.7.5.3.  Import volumes are not indicative of a significant competitive constrain

(936)    The Commission observes that the volumes of HDG imports and their share of the market is notably smaller than that for  HR:  While  for  HR  imports  represented over [20-30]% of the sales market share in 2016, for HDG the figure remained         at [10-20]%. Unlike imports for HR, which have a significantly higher sales market share in Southern Europe ([30-40]%) than in the EEA as a whole, imports of HDG are more evenly distributed across the EEA: their market share in Southern Europe remained  at  [10-20]%  in  2016.   In   Italy  the   HDG  import  market  share  was   at [10-20]%, in contrast to the figure for HR imports which was [30-40]%.  Therefore, the Commission considers that the role of non-EEA producers and imported HDG in the EEA and, in particular in Southern Europe region – including  in Italy– is more limited than that for HR.

(937)    The Commission observes that, as suggested by the Notifying Party, volumes of HDG imports have been on the rise in recent years. The development of the volume of HDG imports in past years is shown in Figure 105.

(938)    A similar development appears to have been on-going in Southern Europe. Figure 106 shows the development of HDG imports in Italy, which is the main steel consumption hub in Southern Europe.

(939)    Further, the Commission notes that imports of HDG in the second part of 2017 have reduced as compared to the first part of 2017 by a factor of [20-30]%. On the  contrary in year 2016 imports of HDG grew by [0-5]% between the first and the second half of the year (Figure 107). Therefore, on this basis, despite the relative increase in HDG imports in the first part of 2017, the most recent trends indicates an attenuation of the inflow of HDG imports into the EEA market.

(940)    Nonetheless, the Commission considers that despite the growth in HDG import volumes both Ilva and ArcelorMittal were able to increase their price premium over the import prices for similar HDG products. As shown in Figure 105 HDG imports increased significantly from 2014 to 2015, at the same time the price premium over imports of ArcelorMittal and Ilva also increased significantly (Figure 108). This indicates that the higher import inflow of HDG was not sufficient to counter a  relative price increase of domestic EEA producers.

Figure 108 – Price difference over Southern European import price

[…]

Source: Commission's computation on the basis of Parties' data and SBB data provided by the parties for HDG southern import price

(941)    Furthermore, the Commission recalls that a number of trade defence measures have been enacted by the European Commission with regard to HDG products, as elaborated in Section 5.3.3.

(942)    In particular, on 9 December 2016, the Commission initiated an anti-dumping proceeding concerning imports of certain corrosion resistant steels (including HDG) originating in China. On 7 July 2017, those corrosion resistant steels have been made subject to registration. (660) Subsequently, on 9 August 2017 the Commission imposed provisional anti-dumping duties on those products. (661) The provisional duties  are listed in Table 51. Definitive measures followed on 8 February 2018 under case number AD639. (662)

(943)    The Commission observes in this respect that import volumes and market shares of non-EEA producers for 2016 do not take into account the effects of these trade defence measures in 2017 and in the near future.

(944)    Moreover, respondents to the market investigation considered that their effects on HDG imports are already seen in the market: The majority of customers taking a position indicated that the pressure from imports in HDG has diminished in the  period between 2014 and 2017. (663) A customer explains: ‘Duties have reduced pressure on the market’. (664) Overall, customers considered the pressure from  imported HDG to be moderate if not limited with particularly industrial customers seeing the pressure as small. (665)

9.7.5.4. Imports can only contest a part of the demand of EEA customers

(945)    The Commission notes that the ability of imports to reliably and continuously serve the EEA market is not only limited by their focus on domestic sales, but also by the characteristics of the demand of EEA customers: The Commission's investigation indicates that only a limited portion of HDG demand in the EEA could be challenged by imports.

(946)    On the basis of ArcelorMittal's […] the Commission estimates that […]% of the EEA customers purchase less that 10% of their HDG needs from imports. These customers account for […]% of the total volumes of HDG purchased in the EEA according to the Notifying Party’s data. More detailed figures are shown in Table 52.

(947)    Looking at Southern Europe, the Commission estimates that […]% of the Southern European customers purchase less that 10% of their HDG needs from imports. These customers account for […]% of the total volumes of HDG purchased in Southern Europe according to the Notifying Party’s data. More detailed figures are shown       in Table 53.

(948)    Furthermore, in the market investigation, a number of customers have indicated that imports could only substitute part of their sourcing from EEA-producers and, most importantly, that they would only switch some, if any, volumes of HDG to non-EEA producers even if prices of EEA supplies saw a small but significant increase relative to imports. Customers named reasons such as lead times, commercial terms and quality as reasons for the phenomenon. (666)

9.7.6. Customers are concerned about the effects of Transaction

(949)    During the market investigation, customers expressed concerns about the effects of the Transaction on HDG. In particular, a majority of distributor customers taking a position expect price increases in HDG due to the Transaction, some quantifying the expected increase as 5% or even 10–15%. A distributor customer notes: ‘Hot-dip- galvanised is already difficult to acquire, even from import’ (667) while another distributor explains: ‘Since mid-1990s, Ilva has had a calming role in the Italian market for the benefit of end users and their exports. Therefore, acquisition of Ilva by ArcelorMittal leads us to expect future price increase in Italy.’ (668),  (669)  As to industrial customers, a significant share (albeit not a majority) of them also expect price increases  due  to  the  Transaction,  a  few  quantifying  their  expectation  as  5% or 10%. (670)

(950)    This finding is corroborated by the fact that a number of customers, particularly industrial customers, indicated in the market investigation that they would have no other choice but to pay a small but significant price increase if ArcelorMittal requested it – even without a reference to the Transaction. An industrial customer explained that ‘when arcelor increase their prices, all suppliers make the same!’. A distributor concurs: ‘If ArcerlorMittal increases the price, other EEA mills will follow.’ (671)

9.7.7. Conclusion on horizontal non-coordinated effects in the market for the production and supply of HDG flat carbon steel products

(951)    In conclusion, based on the evidence available to the Notifying Party, the Commission considers that the Transaction is likely to significantly impede effective competition through the removal of an important competitive constraint in the market for the production and supply of HDG products in the EEA as a result of horizontal non-coordinated effects. For the reasons set out in Section 9.6, the Commission reaches the same conclusion for GS as a whole.

9.8. Production and supply of electro galvanised flat carbon steel products (EG): horizontal non-coordinated effects

9.8.1. The merged entity will be the largest EG supplier, with increased market power, in the EEA

(952)    Prior to the Transaction, ArcelorMittal is already the market leader in EG in terms of sales market share and capacity share under all alternative geographic market definitions.

(953)    At the EEA-level, ArcelorMittal had a sales market share of [30-40]% and a capacity share of [40-50]% % in 2016. Combined the Parties achieved a sales market share of [30-40]% and a capacity share of [50-60]%. The largest competitor, ThyssenKrupp, had a sales market share of [20-30]%. The only other EEA supplier  to achieve a 10% sales market share  at  the  EEA  level  in  2016  was  Salzgitter  with [10-20]% of the merchant market. Imports remained in total at [5-10]%.

(954)    In  Southern  Europe  region (672),  ArcelorMittal  is  also  the  market  leader  with a [50-60]% sales market share in 2016. Combined the Parties achieved a combined sales  market  share   of   [50-60]%.  The  largest  competitor,  ThyssenKrupp,  had a [20-30]% sales market share. No other EEA producer reached 10% sales share in Southern Europe. The import market share remained at [0-5]% in Southern Europe   in 2016.

(955)    The Parties’ and their competitors’ sales market shares are shown in Table 54. The Parties’ capacity shares are shown in Table 55.

(956)    The Notifying Party submitted that, despite the Parties’ high combined market  shares, the market share increment is small and not indicative of competition concerns.

(957)    Nonetheless, the Commission observes that ArcelorMittal has already prior to the Transaction a strong market position for EG in the EEA and, even more so, in Southern Europe region where its market share is also significantly higher than for HDG. Compared to HDG, ThyssenKrupp has a notable market share but overall the number of competitors in EG is limited and smaller than in HDG. The market share increment also appears to be significant in this context.

(958)    The Transaction will increase ArcelorMittal’s market position further with all remaining competitors being significantly smaller both in terms of sales market share and capacity share. Therefore, the Commission considers that the merged entity is likely to have increased market power both at the EEA-level and in particular in Southern Europe.

9.8.2. Other conditions for competition

(959)    The  Commission  observes  that,  in  general,  the  findings  for  HDG  explained     in Section 9.7 are largely applicable to EG as well, particularly with respect to commodity EG. In general the findings are indicative of current competitive interaction between the Parties, the role of Ilva as an important competitive force and the lack of adequate competitive constraints from other EEA producers and imports following the Transaction.

(960)    First, the results of the market investigation indicate that the Parties are close competitors to each other at least with regard to commodity EG within the EEA.

(961)    Accordingly, the majority of customers taking a position considered Ilva and ArcelorMittal to be close competitors in terms of price of EG commodity products. With regard to the quality of EG commodity products, the majority of distributor customers taking a position considered that the Parties compete closely while industrial customers were split on the matter. Some customers even considered the Parties to compete closely in high-end EG products but industrial customers mostly disagreed with such view. (677)

(962)    With regard to geographic closeness between the Parties, it can be noted that the overwhelming majority of distributors that took a position considered in the market investigation that the Parties are competing closely in commodity EG in Italy and its surroundings. (678) The same indication was given by a majority of industrial customers taking a position (even if not as overwhelming as for distributor customers). (679)

(963)    Second, while competitors have some spare capacities, they do not seem to curb the Parties high market share particularly in Southern Europe.

(964)    To this effect, the Commission notes that, according to the results of the market investigation, the Parties’ competitors had  a  readily  available  spare  capacity  in  the EEA in 2016 of […] kt and, considering spare capacities that can be put to use in the short/medium term, […] kt. The Commission observes in this respect that the competitors’ spare capacities are nonetheless not insignificant compared to the sales of the Parties: In 2016, Ilva sold approximately […] kt and ArcelorMittal approximately […] kt of EG in the EEA.

(965)    Nonetheless, the Commission observes that despite the spare capacities of competitors, they have not acted as limits on the Parties’ market shares, particularly not in Southern Europe region.

(966)    Third, imports appear not to exert a significant competitive constraint in EG. The Commission recalls in this respect that there are relatively limited imports of EG into the EG, including into Southern Europe. As shown in Table 54, imports only constituted [5-10]% of the merchant market in the EEA and [0-5]% in Southern Europe. The import level is thus significantly lower than for HDG.

(967)    While there has been variation between years, the import volumes for EG have been decreasing for the past decade. In line with this, imports in 2017 were still lower than in 2014 and a number of years before that even if they had increased somewhat compared to 2016 and 2015. Figure 109 shows the development of EG imports into the EEA during the past years.

(968)    Considering imports of EG into Southern Europe, Figure 110 shows the import volumes into Italy, which is the main steel consumption hub in Southern Europe. As seen in the Figure, the volumes have in general been rather small and volatile, and no long-standing trend of increased imports can be observed.

(969)    During the market investigation, some customers also explicitly noted the absence of competitive constraint from imports with regard to EG. An industrial customer notes: ‘[m]arginal role for EG material compared to HDG’ while a distributor explains that‘EG is for more dedicated products. Difficult to see it from import offers’. (680)

(970)    Fourth, customers are concerned about the Transaction with regard to EG. A  majority of both industrial and distributor customers taking a position expect price increases due to the Transaction. An industrial customer explains: ‘ArcelorMittal will become even greater dominant position and will extinguish another competitor in Europe [sic]’ while a distributor concurs: ‘In a situation of almost monopoly, the price will be influenced by the new entity’ (681). (682)

9.8.3.Conclusion on horizontal non-coordinated effects in the market for the production and supply of EG flat carbon steel products

(971)    In conclusion, based on the evidence available to it, the Commission considers that the Transaction is likely to significantly impede effective competition through the removal of an important competitive constraint in the market for the production and supply of electro-galvanised steel in the EEA as a result of horizontal non- coordinated effects.

9.9. Production and supply of metallic coated steel for packaging

9.9.1. The Notifying Party's views

(972)    The Notifying Party submitted that the Transaction does not give rise to competition concerns related to horizontal overlaps in the market for metallic coated steel for packaging, that is TP and ECCS because (i) the Parties' combined market shares are low with a small increment; (ii) the Parties are not close competitors; (iii) the merged entity will continue facing competition from EEA competitors, in particular ThyssenKrupp and Tata Steel; (iv) imports from outside the EEA will continue to provide a constraint on the merged entity after the Transaction, in particular in Italy; and (v) the Parties' customers have significant buyer power and can easily switch suppliers.

9.9.2.The Commission's assessment

(973)    ArcelorMittal and Ilva are both active in the production and supply of metallic coated steel for packaging, including both TP and ECCS. In its competitive assessment of the Transaction as regards the market for the production and supply of metallic coated steel for packaging in the EEA, the Commission makes the following observations.

(974)    First, the Parties' combined sales market share for metallic coated steel for packaging in the EEA in 2016 represented [30-40]% with a small increment of [0-5] percentage points, as depicted in Table 56. (683)

(975)    The Parties' combined sales market  share  for  TP  in  the  EEA  in  2016  represented [20-30]% with an increment of [0-5]% percentage points while their combined share for ECCS was [40-50]% with an increment of [5-10] percentage points. The latter, however, represents only […] kt of sales volumes. (685)

(976)    Table 57 shows the sales shares in Italy. The merged entity would have sales shares of [20-30]% in metallic coated steel, with an increment of [10-20]%. In TP, the merged entity would have sales shares of [20-30]%, with an increment of [5-10] percentage points. Finally, in ECCS, the merged entity would have sales shares of [30-40]%, with an increment of [10-20] percentage points. The latter, however, represents only […] kt of sales volumes.

(977)    Second, in addition to the limited volumes of TP and ECCS sold by Ilva, the Commission notes that the structure of the EEA market for metallic coated steel for packaging is different than in the case of HR, CR and GS. In particular, ThyssenKrupp and Tata Steel also have significant market shares ([20-30]% and [10-20]% in 2015 respectively) that are comparable to those of ArcelorMittal both pre- and post-Transaction at EEA level. ThyssenKrupp and Tata Steel also have substantial sales of TP in Italy ([20-30]% and [10-20]% in 2015 respectively). Whilst in Italy, Tata Steel's sales of ECCS ([0-5]% in 2015) are not comparable to those of ArcelorMittal, Ilva and ThyssenKrupp ([10-20]% in 2015), the overall market size in terms of sales of ECCS in Italy in 2016 was only […] kt with imports taking [40-50]% of the volumes sold. (687) As regards imports, the Commission observes that unlike in the case of HR, CR and GS, there are currently no anti-dumping measures imposed by the Commission on imported TP or ECCS.

(978)    Thus, the market structure suggests that post-Transaction the merged entity would face competition from other large EEA steel producers, notably ThyssenKrupp and Tata Steel, as well as from imports from non-EEA producers in the market for the production and supply of metallic coated steel for packaging (including the potential sub-segments TP and ECCS), both on an EEA-wide level and in Italy. (688)

(979)    Third, the feedback from the Commission's market investigation suggests that the competitive interaction and closeness of competition between ArcelorMittal and Ilva is less apparent than in the case of HR, CR and GS. The Commission's market investigation does not indicate that the Parties would be close competitors in the production and supply of TP nor ECCS. One customer explained that 'ArcelorMittal and Ilva serve slightly different purposes in [the company's] purchasing strategy, ArcelorMittal being the development and quality "partner" while Ilva provides for good price.' (689) Another large customer noted that 'Ilva mainly supplies the standard commodity grades, also known as "tomato specifications".' (690) In fact, the  Commission observes that ThyssenKrupp and Tata Steel seem to compete more closely with ArcelorMittal than Ilva and imports seem to be the closest competitor of Ilva in TP and ECCS. A large customer of TP noted that imports are particularly present on the commodity grades manufactured by Ilva and particularly in Italy. (691) Another customer that sources both TP and ECCS made a remark that 'the main competitive dynamic in Europe is between ArcelorMittal, Tata Steel and ThyssenKrupp Rasselstein.' (692)

(980)    Finally, only a limited number of customers that responded to the Commission's investigation indicated that they would expect a price increase in the prices for TP and ECCS as a result of the Transaction. (693) In line with that feedback and as  discussed above, it is unlikely that the merged entity would have the ability and incentive to increase the prices for TP and ECCS post-Transaction, given the constraint other large EEA producers, ThyssenKrupp and Tata Steel, would exert on the merged entity. As such, the risk of price increases for metallic coated steel for packaging post-Transaction is limited and certainly lower than for the other flat carbon steel products, namely HR, CR and GS.

9.9.3. Conclusion on horizontal non-coordinated effects in metallic coated steel for packaging

(981)    For the reasons explained in this Section, and considering the evidence available to  it, the Commission finds that the Transaction would not result in a significant impediment to effective competition due to horizontal non-coordinated effects in the market for the production and supply of metallic coated steel for packaging,  including its potential sub-segments.

9.10. Production and supply of welded carbon steel tubes

9.10.1. Small welded carbon steel tubes

(982)    ArcelorMittal and Ilva are both active in the production and supply of small welded carbon steel tubes.

9.10.1.1.The Notifying Party’s views

(983)    The Notifying Party submitted that the Transaction does not give rise to competition concerns related to horizontal overlaps in the markets for small welded carbon tubes because the Parties' combined market shares are low, there is substantial  overcapacity as regards small welded carbon steel tubes and a number of alternative suppliers will remain active. In addition, the Parties are not close competitors, as ArcelorMittal primarily supplies tubes to the automotive and construction industries, whereas Ilva focuses primarily on other industries, with limited supplies to the construction industry. Furthermore, it is submitted that barriers to entry are low and investment for the necessary machinery and knowhow is easily obtainable in this market segment.

(984)    The Notifying Party also submitted that combined market shares in the EEA remain below [10-20]%, regardless of product market definition. (694) On a national level, horizontally affected markets do arise in the segments of small welded carbon steel tubes in France and Italy. However, even in these narrower markets, the Parties' combined market shares remain below 30%, however the market is defined.

(985)    Furthermore, the Notifying Party indicated that post-Transaction many other strong competitors will remain active in the production and supply of small welded carbon steel tubes on the Italian market).

9.10.1.2.The Commission's assessment

(986)    ArcelorMittal and Ilva are both active in the production and supply of small welded carbon tubes. In its competitive assessment of the Transaction as regards the market for the production and supply of small welded carbon tubes in the EEA, the Commission makes the following observations.

(987)    First, the Parties' combined sales  market share  for small welded  carbon tubes in   the EEA in 2016 represented [5-10]% with a small increment of [0-5] percentage points, as depicted in Table 58. On the narrower product markets of small welded carbon standard tubes and cold-finished small welded carbon tubes, the Parties' combined  sales  market  shares  in  the  EEA  in  2016  amounted  to  [10-20]% and [5-10]%, with a small increments of [0-5]% and [0-5]% respectively. As such, the EEA market in these products is not affected as such. (695)

(988)    On a national level, whilst horizontally affected markets do arise in the segments of small welded carbon steel tubes in Italy and France, even in these narrower markets, the Parties' combined market shares remain below 30%, however the market is defined.

(989)    Table 59 shows the sales shares in Italy, where the Parties' activities overlap in small welded standard tubes. There, based on 2016, the merged entity would have sales shares of [20-30]%, with a minor increment of [0-5]%.

(990)    Table 60 shows the sales shares in France. There, based on 2016, the merged entity would have a combined market share of [20-30]%, with an increment of [5-10]%.

(991)    The majority of respondents to the market investigation indicated that  the Transaction would not adversely affect competition in any significant manner. Most respondents indicated that the Transaction would not affect their ability to source small welded carbon steel tubes in the EEA. Furthermore, a majority of both customers and competitors indicated that they do not expect that the Transaction would lead to an increase  in  the  price  for  small  welded  carbon  steel  tubes  in  the EEA. (696) Only a small minority of customers and of competitors who responded expect a price increase as a result of the Transaction. (697) Most expected that post- Transaction, the price would either remain the same or decrease.

(992)    Some respondents to the market investigation raised concerns that, post-Transaction, ArcelorMittal could have a competitive advantage over its competitors due to its size, its nature as a vertically integrated supplier and an ability to leverage Ilva's production facilities in Italy. (698)

(993)    However, the Commission notes in this regard that post-Transaction a number of competitors would remain active in the Italian market for small welded carbon steel tubes (for example Marcegaglia [40-50]%, Arvedi [5-10]–[10-20]%, Padana [10-20]–[20-30]%,   Profiltubi [5-10]–[10-20]%   and   Tecnotubi  [5-10]–[10-20]%).

This is reflected by the feedback of those customers who responded that the Transaction would not lead to higher prices because there are a sufficient number of alternative suppliers active in the market. (699)

(994)    Furthermore, as shown at Table 60, post-Transaction a number of competitors would also    remain    active    in    Franch    for    small    welded    tubes    (for example Marcegaglia [10-20]%, Arvedi [0-5]%, Padana [10-20]%, Tata  [0-5]% and  Tecnotubi [0-5]%.).

9.10.2.   Conclusion on horizontal non-coordinated effects in the production and supply of small welded carbon steel tubes

(995)    For the reasons explained in this Section, and considering the evidence available to  it, the Commission finds that the Transaction would not result in a significant impediment to effective competition due to horizontal non-coordinated effects with respect to the market for the production and supply of small welded carbon steel tubes.

9.10.3.   Large welded carbon steel tubes

(996)    Both Parties are active in the manufacturing of longitudinally submerged arc welded tubes (‘LSAW’) and spirally welded tubes.

9.10.3.1. The Notifying Party’s view

(997)    The large welded tubes segment is project driven on an international scale. As such the Notifying Party was not able to provide sales estimates for competitors. However it was able to provide estimates of the capacity shares of the Parties' main LSAW competitors in the EEA, as well as tubes manufacturers in Turkey and Macedonia which compete in the EEA market. Furthermore, the Notifying Party has submitted market shares on the basis of the EEA only. The Commission has found this to be an acceptable proxy in the context of the case.

(998)    The Notifying Party submitted that as the Parties' combined market shares are very minor and as the Parties face competition from well-established competitors for both LSAW and spirally welded tube, it is inconceivable that the Transaction would have any significant impact in this area.

9.10.3.2. The Commission's assessment

(999)    ArcelorMittal and  Ilva are both active in the production and supply of LSAW in    the EEA. In its competitive assessment of the Transaction as regards the market for the production and supply of LSAW in the EEA, the Commission makes the following observations.

(1000)   As depicted in Table 61, the Parties' combined sales market share for LSAW in       the EEA in 2016 represented just [0-5]% with a minor increment of [0-5]%. As such, the EEA market in LSAW is not affected.

(1001) Post-Transaction many other competitors will remain active in the production and supply of LSAW both in the EEA and outside of it. These include Europipe (Germany), Khartszysk (Ukraine), Corinth Pipe Works (Greece), TATA Steel (United Kingdom) and Erndtebruckner Eisenwerk (Germany).

(1002) Similarly, the Parties’ combined market share remains clearly below 20% and the market share increments very limited if considering large spirally welded carbon  steel tubes in isolation, as shown in Table 62. As such, the EEA market for large spirally welded carbon steel tubes is not affected.

(1003) Post-Transaction many other competitors will remain active in the production and supply of spirally welded tubes both in the EEA and outside of it. The Parties' main competitors in the EEA for large spirally welded carbon steel tubes are Corinth Pipe Works (Greece), STS (Spain), Salzgitter Groessrohre (Germany), Noksel (Spain), Bender Ferndorf (Germany) and Hannibal (Spain).

(1004) Therefore, and on the basis of feedback given in the market investigation, (700) it is unlikely that the Transaction would lead to price increases on the EEA market for large welded carbon steel tubes, irrespective of the exact product and geographic market definition.

9.10.4. Conclusion on non-coordinated effects as regards the production and supply of large welded carbon steel tubes

(1005)   For the reasons explained in this Section, and considering the evidence available to   it, the Commission finds that the Transaction would not result in a significant impediment to effective competition due to horizontal non-coordinated effects with respect to the market for the production and supply of large welded carbon steel tubes under any alternative market definition.

9.11. Distribution of flat carbon steel

(1006) The Parties’ activities in distribution of flat carbon steel only overlap with regard to  the distribution of flat carbon steel through SSCs in Italy and France. Indeed, while ArcelorMittal  has  SSCs  in  numerous  countries,  Ilva  has  SSCs  only  in  Italy  and France. (701) The Italian market is not affected  as the combined market  share of  the Parties remains  below 20%.  The Parties' market  shares  in  France  are shown  in Table 63. (702),  (703)

(1007) The Commission observes that while the Parties’ combined market share in France is not insignificant, the market share increment remains small at only [0-5]%. Furthermore, there are numerous competitors such as Tata, Robert and Raymond GANS as well as a large number of smaller players.

(1008) The Commission further observes that the increment brought  about  by  the Transaction would in any event be addressed by the remedy proposed by the Notifying Party.

(1009) When considering a distinct market for the distribution of TP through SSCs, as suggested by the Notifying Party, the Commission observes that there is no  horizontal overlap, as Ilva is not active in this segment.

(1010) Therefore, the Commission finds that the Transaction would not significantly impede effective competition in the internal market with respect to horizontal non- coordinated effects in the market of steel distribution through SSCs in any country in the EEA.

9.12. Vertical links between semi-finished and finished flat carbon steel products

9.12.1.    Structure of the markets

(1011)   The  Transaction  gives  rise  to  a  vertically  affected  market  between  the   Parties’

(i)   upstream  production  and  supply of  semi-finished  flat  carbon  steel  (slabs) and

(ii)   their downstream production and supply of HR products in the EEA.

(1012)  In the upstream markets, ArcelorMittal and Ilva have significant capacities, which  can be approximated by their capacity for HR products. As regards merchant sales, however, only ArcelorMittal is practically present. While the Parties both produce slabs, only ArcelorMittal regularly sells meaningful amounts to third parties, (704) but  its market share remains below 30% under any market definition (705).

(1013) In the downstream markets, the Parties have a combined market share of over 30% in Southern Europe as shown in Table 24, and a capacity share over 30% at the EEA level as shown in Table 22.

9.12.2. Input foreclosure

(1014) The Commission observes that, different from HR, semi-finished products / slabs are sold less on the merchant market by the Parties. Moreover, as far as input foreclosure is concerned, major HR suppliers such as ThyssenKrupp, Tata, Salzgitter, Voestalpine, SSAB and Arvedi have internal slab production that covers all or at  least the major share of their needs. This reduces the likelihood of effects on the downstream HR market of any input foreclosure.

(1015) During the market investigation, no concerns were raised with regard to merger- specific input foreclosure of slabs.

(1016) Therefore, the Commission finds that concerns of input foreclosure in relation to semi-finished flat carbon steel products (slabs) are unlikely to arise as a result of the Transaction.

9.12.3.Customer foreclosure

(1017) The Parties are both integrated producers of flat carbon steel, and they primarily satisfy their needs for slabs through internal production. Even if they from time to time purchased limited volumes of slabs for instance because of unforeseen maintenance works at their steel-making facilities, they are not significant customers for slabs in the EEA. This also applies for Ilva that has, despite the current  limitations on its upstream production, not significantly resorted to purchasing slabs for its downstream operations.

(1018) Consequently, the Commission considers that concerns of customer foreclosure in relation to semi-finished flat carbon steel products (slabs) are unlikely to arise as a result of the Transaction.

(1019) In any event, the Commission also observes that the proposed remedies encompass integrated facilities with own slab manufacturing capabilities.

9.12.4. Conclusion on the vertical links between the production and supply of semi-finished and finished flat carbon steel products

(1020) The Commission therefore finds that the Transaction would not lead to significant impediment to effective competition in the internal market as a result of vertical non- coordinated effects with regard to the vertical links between the production and supply of semi-finished and of finished flat carbon steel products.

9.13. Vertical links between the production and supply of finished flat carbon steel products

9.13.1. Introduction

(1021) The Transaction gives rise to vertically affected markets within the Parties’ activities  in the flat carbon steel value chain in the EEA, in particular between:

(a)   HR and its downstream products CR and HDG (similarly if an overall GS market is considered);

(b)   CR and its downstream products HDG and EG (or overall GS), metallic coated steel for packaging and OC; as well as

(c)   HDG and EG (or overall GS), and their downstream product OC.

(1022) Both of the Parties produce and supply all of the products discussed with the  exception of OC that only ArcelorMittal produces and supplies.

9.13.2.Vertical links between HR and its downstream products

9.13.2.1.Structure of the markets

(1023) The Parties both produce HR and they also use it as an input in their production and supply of CR and HDG. (706)

(1024) As shown in Table 24, the Parties’ combined sales share exceeds 30% in Southern Europe for HR. As shown in Table 22, the Parties’ combined upstream capacity exceeds 30% at the EEA-level.

(1025) In the downstream markets for CR and HDG, the Parties’ combined market shares exceed 30% both in Southern Europe as well as at EEA-level. As shown in Table 22, concerning the EEA-capacity, the Parties’ combined capacity shares also reach 30% both in CR and HDG.

(1026) The findings are the same if an overall GS market is considered instead of an HDG only market.

9.13.2.2.Input foreclosure

(1027)   As discussed in Section 9.4, the merged entity would have a strong market position   in the upstream market for the production and supply of HR in the EEA and the Transaction is likely to significantly impede effective competition as a result of horizontal non-coordinated effects in that market. As explained in Section 10, the Commission also finds that the Transaction may significantly impede effective competition due to horizontal coordinated effects in the production and supply of HR in the EEA. Furthermore, the merged entity would have a notable market presence in the downstream markets for CR and HDG.

(1028) Hence, a question arises whether the merged entity could foreclose its downstream rivals from HR inputs, for example by increasing prices it charges to them.

(1029) Based on the evidence available to it, the Commission considers that any such ability  of the merged entity, and in particular any increase thereto, would be a consequence of the horizontal non-coordinated and coordinated effects in the production and supply of HR as described in this Decision. Similarly, any increase in the incentive  of the merged entity to foreclose – compared to the Parties separately – would be a consequence of the overlap in the downstream markets.

(1030) As discussed in detail in Section 13, the commitments entered into by the Notifying Party significantly reduce the overlap brought by the Transaction on both upstream and downstream markets. The Commitments are also sufficient to address the Commission’s horizontal non-coordinated and coordinated competition concerns in those markets. The Commission hence considers that the commitments would in any event remove any merger-specific ability for the merged entity to engage in input foreclosure.

(1031) Therefore, the Commission considers that, in view of the final commitments entered into by the Notifying Party, it is not necessary to conclude on input foreclosure as the commitments would in any event address such potential concerns.

9.13.2.3.Customer foreclosure

(1032) The Parties are both integrated producers of flat carbon steel, and they primarily  satisfy their needs for HR through internal production. Even if they from time to time purchased limited volumes of HR for instance because of unforeseen maintenance works at their HR facilities, they are by no means important customers for HR in    the EEA. This also applies for Ilva that has, despite the current limitations on its upstream production, not significantly resorted to purchasing HR for its downstream operation but has rather directed HR from merchant sales to satisfy its internal needs.

(1033) Consequently, the Commission considers that the Transaction  does  not  raise  concerns in relation to customer foreclosure.

9.13.3. Vertical links between CR and its downstream products

9.13.3.1.Structure of the markets

(1034) The Parties both produce CR and they also use it as an input in their production and supply of (i) HDG, (ii) EG and (iii) metallic coated steel for packaging (TP/ECCS). ArcelorMittal also produces and supplies (iv) OC that can be produced using CR as the substrate.

(1035) As shown in Table 21 and Table 24, the Parties’ combined upstream market share in CR reaches 30% both at the EEA-level as well as in Southern Europe. As shown      in Table 22, concerning EEA-capacity, the Parties’ combined upstream capacity also reaches 30%.

(1036)  With regard to the downstream products, the Parties have high market shares that   also exceed 30% for HDG and EG as well as in the potential overall market for GS as shown in Table 21 and Table 24. For all of the other products concerned, the information provided by the Notifying Party shows that the Parties’ combined market shares reach 30% for all of them, including OC, with the exception of certain sub- segments of metallic coated steel for packaging where the Parties’ combined market share remains below 30% (for instance TP in the EEA [20-30]%). However, even those potential sub-segments are vertically affected due to the Parties’ upstream market shares in CR.

9.13.3.2.Input foreclosure

(1037)   As discussed in Section 9.5, the merged entity would have a strong market position   in the upstream market for the production and supply of CR in the EEA and the Transaction is likely to significantly impede effective competition due to horizontal non-coordinated effects in that market. As explained in Section 10, the Commission also finds that the Transaction may significantly impede effective competition due to horizontal coordinated effects in the production and supply of CR in the EEA. Furthermore, the merged entity would have a notable market presence in the downstream markets for HDG and EG.

(1038) Hence, a question arises whether the merged entity could foreclose its downstream rivals, particularly those for HDG and EG, from CR inputs, for example by increasing prices it charges to them.

(1039) Based on the evidence available to it, the Commission considers that any such ability  of the merged entity, and in particular any increase thereto, would be a consequence of the horizontal non-coordinated and coordinated effects in the production and supply of CR as described in this Decision. Similarly, any increase in the incentive of the merged entity to foreclose – compared to the Parties separately – would be a consequence of the overlap in the downstream markets.

(1040) As discussed in detail in Section 13, the commitments entered into by the Notifying Party significantly reduce the overlap brought by the Transaction on the upstream market for CR, and they reduce or even remove the overlaps in the downstream markets for HDG and EG. The Commitments are also sufficient to address the Commission’s horizontal non-coordinated and coordinated competition concerns in those markets. The Commission hence considers that the commitments would in any event remove any merger-specific ability for the merged entity to engage in input foreclosure.

(1041) Therefore, the Commission considers that, in view of the final commitments entered into by the Notifying Party, it is not necessary to conclude on input foreclosure as the commitments would in any event address such potential concerns.

9.13.3.3.Customer foreclosure

(1042) The Parties are both integrated producers of flat carbon steel, and they primarily satisfy their needs for CR through internal production. Even if they from time to time purchased limited volumes of CR for instance because of unforeseen maintenance works at their CR facilities, they are by no means important customers  for CR in   the EEA. This also applies for Ilva that has, despite the current limitations on its upstream production of HR (and consequently CR), not significantly resorted to purchasing CR for its downstream galvanising operation but has rather directed material flows from merchant sales to satisfy its internal needs.

(1043) Consequently, the Commission considers that the Transaction  does  not  raise  concerns in relation to customer foreclosure.

9.13.4. Vertical links between HDG/EG and their downstream products

9.13.4.1.Structure of the markets

(1044)   The Parties both produce galvanised steel (HDG and EG). ArcelorMittal, but not   Ilva, produces OC. (707)

(1045) As shown in Table 21 and Table 24, the Parties’ combined upstream market share in HDG and EG reaches 30% both at the EEA-level as well as in Southern Europe. As shown in Table 22, capacity-wise the finding is similar. The findings are similar if considering an overall GS market instead of distinct HDG and EG markets.

(1046) As to the downstream OC market, information provided  by the  Notifying  Party shows that ArcelorMittal’s market share in OC also exceeds 30% in the EEA as well as in Southern Europe.

9.13.4.2.Input foreclosure

(1047) As discussed in Sections 9.6 to 9.8, the merged entity would have a strong market position in the upstream market for the production and supply of HDG and EG, as well as in the potential overall market for the production and supply of GS in the EEA, and the Transaction is likely to significantly impede effective competition in those markets as a result of horizontal non-coordinated effects. As explained in Section 10, the Commission also finds that the Transaction may significantly impede effective competition due to horizontal coordinated effects in the production and supply of HDG and EG (and in the potential overall market for the production and supply of GS) in the EEA.

(1048) Hence, a question arises whether the merged entity could foreclose its downstream rivals from HDG and EG inputs, for example by increasing prices it charges to them.

(1049) Based on the evidence available to it, the Commission considers that any such ability  of the merged entity, and in particular any increase thereto, would be a consequence of the horizontal non-coordinated and coordinated effects in the production and supply of HDG and EG as described in this Decision.

(1050) In addition, the Commission notes that the Transaction does not result in any overlap on the downstream markets for OC, suggesting that the Transaction does not increase the incentive of ArcelorMittal to engage in input foreclosure.

(1051)  Furthermore, as discussed in detail in Section 13, the commitments entered into by  the Notifying Party significantly reduce or even remove the overlap brought by the Transaction on the upstream markets. The Commitments are also sufficient to  address the Commission’s horizontal non-coordinated and coordinated competition concerns in those markets. The Commission hence considers that the commitments would in any event remove any merger-specific ability for the merged entity to engage in input foreclosure.

(1052) Therefore, the Commission considers that, in view of the final commitments entered into by the Notifying Party, it is not necessary to conclude on input foreclosure as the commitments would in any event address such potential concerns.

9.13.4.3.Customer foreclosure

(1053) ArcelorMittal is an integrated producer of flat carbon steel, and it primarily satisfies  its needs for HDG and EG through internal production. Even if it from time to time purchased limited volumes of HDG or EG for instance because of unforeseen maintenance works at its galvanising facilities, it is by no means an important customer for HDG or EG in the EEA.

(1054) Consequently, the Commission considers that the Transaction  does  not  raise  concerns in relation to customer foreclosure.

9.13.5. Conclusion on vertical links between the production and supply of finished flat carbon steel products

(1055) The Commission finds that, in view of the final commitments entered into by the Notifying Party, it is not necessary to conclude as to whether the Transaction would lead to a significant impediment to effective competition in the internal market due to input foreclosure with regard to the vertical links between the production and supply of finished flat carbon steel products, as the commitments would in any  event address such potential concerns.

(1056) The Commission further considers that the Transaction would not lead  to  a  significant impediment to effective competition in the internal market due to customer foreclosure with regard to the vertical links between the production and supply of finished flat carbon steel products.

9.14. Vertical links between the production and supply of finished flat carbon steel products and steel distribution

9.14.1.        Structure of the markets

(1057)   The Transaction results in a vertical link and vertically affected markets between

(i)   the Parties’ production and supply of various finished flat carbon steel products (upstream) and (ii) the Parties’ activities in the distribution of such steel products (downstream). At the distribution level affected markets arise with regard to both SSCs and SCs.

(1058) As far as merchant sales shares are concerned, as observed in Table 21 and Table 24, the Parties’ combined market share reaches or exceeds 30% in CR, HDG and EG at the EEA-level and also in Southern Europe. For HR, the combined market share reached 30% only in Southern Europe. Information provided by the Notifying Party further show that the Parties’ combined upstream market shares also reach 30% for OC, which only ArcelorMittal produces. The finding is similar for metallic coated steel for packaging with the exception of certain potential sub-segments (for instance TP in the EEA [20-30]%).

(1059)  The Parties’ downstream market shares in the EEA countries where they are active  are included in Table 64.

(1060) The Notifying Party has suggested two alternative market definitions: (i) distribution  of flat carbon steel through SCs encompassing all types of steel and (ii) distribution  of metallic coated steel for packaging through SSCs. The market information provided by the Notifying Party shows that in the alternative SC markets, where Ilva is not present, ArcelorMittal’s market shares are somewhat different from those shown in Table 64 but that  the  highest  market  share  remains  at  [20-30]%  (France, 2016). In the potential markets for the distribution of metallic coated steel for packaging through specialised SSCs, where Ilva is not present, ArcelorMittal’s market shares are somewhat higher, reaching [30-40]% in Italy and [30-40]% in Spain.

9.14.2. The Notifying Party’s arguments

(1061) The Notifying Party submitted that the vertical links between the Parties’ upstream production of flat carbon steels and their downstream distribution of such steel products do not give rise to competition concerns. In particular, the Notifying Party submitted that the merged entity lacks both the ability and incentive to engage into input foreclosure or customer foreclosure. According to the Notifying Party, downstream competitors will continue to have access to sufficient alternative upstream suppliers while those upstream suppliers will also continue to have access to a sufficient customer base.

9.14.3. Input foreclosure

(1062) The Commission notes that the Parties’ combined market shares in the upstream markets are not negligible and the market share increments are also noticeable for HR, CR, HDG and EG. As explained more in detail in Sections 9.4 to 9.8, the Commission has concerns related to horizontal non-coordinated effects and, as explained in Section 10, it also finds that the Transaction may significantly impede effective competition due to horizontal coordinated effects in the production and supply of various flat carbon steel products.

(1063) Hence, a question arises whether the merged entity could foreclose its downstream rivals from access to flat carbon steel products, for example by increasing prices it charges to them.

(1064) Based on the evidence available to it, the Commission considers that any such ability  of the merged entity, and in particular any increase thereto, would be a consequence of the horizontal non-coordinated and coordinated effects in the production and supply of flat carbon steel as described in this Decision.

(1065) As discussed in detail in Section 13, the commitments entered into by the Notifying Party significantly reduce or even remove the overlap brought by the Transaction in the upstream markets. The Commitments are also sufficient to address the Commission’s horizontal coordinated and non-coordinated competition concerns in those markets. The Commission hence considers that the commitments would in any event remove any merger-specific ability for the merged entity to engage in input foreclosure of HR, CR, HDG and EG.

(1066) As regards the market shares for metallic coated steel for packaging, the Commission observes that these exceed 30%. However, as discussed in more detail in Section 9.9, the Commission has no concerns related to the overlap between the Parties’  activities. Taking into account the remaining strong suppliers such as ThyssenKrupp and Tata and all evidence available to it, the Commission further considers that the Transaction is not likely to result in merger-specific ability or incentive for the merged entity to engage in input foreclosure with regard to metallic coated steel for packaging.

(1067) With regard to OC, the Commission observes that the Transaction does not result in horizontal overlaps. The Commission further considers that the limited overlap in the downstream markets does not alone result in input foreclosure concerns due to the Transaction.

9.14.4. Customer foreclosure

(1068) The Commission observes that the downstream market shares shown in Table 64 generally remain modest and are only elevated in SSCs in […] and […]. However, even in those countries, the market share increment brought by the Transaction is small.

(1069) The Commission further observes that the market shares remain modest and the Transaction results in no increment at all in the markets for the distribution of flat carbon steel through SCs, as well as in the potential market for the distribution of all steel through SCs. Similarly, even if the market shares in the potential market for the distribution of metallic coated steel for packaging through specialised SSCs are somewhat higher, the market shares remain relatively modest and the Transaction does not result in any increment.

(1070)   The Commission also notes that the Parties are themselves producers and suppliers   of flat carbon steel, and they satisfy their downstream needs at the distribution level predominantly with their own production. [Commercial  information]. (709)  Furthermore, a significant share of flat carbon steel finds its customers directly in the steel-consuming industry and steel distribution is thus not the only outlet for the products of the Parties’ competitors. Overall, the Parties are by no means important customers for flat carbon steel.

(1071) Consequently, the Commission considers that the Transaction  does  not  raise  concerns in relation to customer foreclosure.

9.14.5. Conclusion on vertical links between the production and supply of flat carbon steels and steel distribution

(1072) The Commission finds that, in view of the final commitments entered into by the Notifying Party, it is not necessary to conclude as to whether the Transaction would lead to a significant impediment to effective competition in the internal market due to input foreclosure with regard to the vertical links between the production and supply of finished flat carbon steel products and the distribution of such products, as the commitments would in any event address such potential concerns.

(1073) The Commission further considers that the Transaction would not lead  to  a  significant impediment to effective competition in the internal market due to customer foreclosure with regard to the vertical links between the production and supply of finished flat carbon steel products, and the distribution of such products.

9.15. Vertical links between the production and supply of finished flat carbon steel products on the one hand and welded carbon steel tubes on the other hand

9.15.1.Structure of the markets

(1074) The Transaction results in vertical links and vertically affected markets between the Parties' production and supply of (i) various finished flat carbon steel products (upstream) on the one hand and (ii) small welded tubes and (iii) large welded tubes on the other hand. According to the Notifying Party, welded carbon steel tubes are mainly made from HR (>[…]%) and less so from CR and GS (<[…]%). (710)

(1075) As observed in Table 21 and Table 24, the Parties’ combined market share reaches or exceeds 30% in CR, HDG and EG at the EEA-level and also in Southern Europe.  For HR, the combined market share reached 30% only in Southern Europe.

(1076) As shown in Table 22, concerning EEA-capacity, the Parties’ combined upstream capacity reached 30% for all of the upstream products concerned.

(1077) For small welded carbon steel tubes, combined market shares in the EEA remain  below [10-20]% regardless of market definition while market shares on a national basis remain below 30%, as described in Table 58, Table 59 and Table 60.

(1078) For large welded tubes, combined market  shares  in  the  EEA  are  also  clearly  below 30%, as described in Table 61 and in Table 62.

9.15.2. The Notifying Party's arguments

(1079) The Notifying Party submitted that the vertical links between the Parties' upstream production and supply of flat carbon steel and the downstream production and supply of small and large welded tubes do not give rise to competition concerns. […]. According to the Notifying Party, downstream competitors will continue to have access to sufficient alternative upstream suppliers while those upstream suppliers  will also continue to have access to a sufficient customer base.

9.15.3. Input foreclosure

(1080) The Commission notes that the Parties’ combined market shares in the upstream markets are not negligible and the market share increments are also noticeable. As explained more in detail in Sections 9.4 to 9.8, the Commission has concerns related to horizontal non-coordinated effects and as explained in Section 10, it also finds that the Transaction may significantly impede effective competition due to horizontal coordinated effects in the production and supply of various flat carbon steel products.

(1081) Hence, a question arises whether the merged entity could foreclose its downstream rivals, for example by increasing prices it charges to them.

(1082) Based on the evidence available to it, the Commission considers that any such ability  of the merged entity, and in particular any increase thereto, would be a consequence of the horizontal non-coordinated and coordinated effects in the production and supply of flat carbon steel as described in this Decision.

(1083) As discussed in detail in Section 13, the commitments entered into by the Notifying Party significantly reduce or even remove the overlap brought by the Transaction in the upstream markets for the production and supply of flat carbon steel products. The Commitments are also sufficient to address the Commission’s horizontal non- coordinated and coordinated competition concerns in those markets. The  Commission hence considers that the commitments would in any event remove any merger-specific ability for the merged entity to engage in input foreclosure.

9.15.4.Customer foreclosure

(1084)   The Commission observes that the downstream market shares remain modest.

(1085) The Commission further notes that the Parties are themselves  producers  and  suppliers of flat carbon steel, and they satisfy their downstream needs at the distribution level predominantly with their own production. Overall, the Parties are by no means important customers for flat carbon steel.

(1086) Consequently, the Commission considers that the Transaction  does  not  raise  concerns in relation to customer foreclosure.

9.15.5. Conclusion on vertical links between the production and supply of finished flat carbon steel products and of welded carbon steel tubes

(1087) The Commission finds that, in view of the final commitments entered into by the Notifying Party, it is not necessary to conclude as to whether the Transaction would lead to a significant impediment to effective competition in the internal market due to input foreclosure concerns with regard to the vertical links between the production and supply of finished flat carbon steel products and of welded carbon steel tubes as the commitments would in any event address such potential concerns.

(1088) The Commission further considers that the Transaction would not lead  to  a  significant impediment to effective competition in the internal market due to customer foreclosure concerns with regard to the vertical links between the production and supply of finished flat carbon steel products and of welded carbon steel tubes.

9.16. Vertical links between the production and supply of finished flat carbon steel products and ArcelorMittal’s activities in various downstream markets

9.16.1. Commission assessment

(1089) In addition to the products mentioned in Sections 9.14 and 9.15, ArcelorMittal is  active in various products that are downstream of the production and supply of finished  flat  carbon  steel  products,  such  as  in  tailor  welded  blanks (711), sandwich panels (712) and in steel construction sheets (713). Ilva has no activities in any of those markets.

(1090) These markets have in common that they use various flat carbon steel products as inputs into their production though the importance of steel as an input varies. In light of the evidence available to it, the Commission considers that the considerations applicable to input foreclosure in the markets discussed in Sections 9.14 and 9.15 also apply to the other vertical links discussed in this section.

(1091) Hence, a question arises whether the merged entity could foreclose its downstream rivals, for example by increasing prices it charges to them.

(1092) Based on the evidence available to it, the Commission considers that any such ability  of the merged entity, and in particular any increase thereto, would be a consequence of the horizontal non-coordinated and coordinated effects in the production and supply of flat carbon steel as described in this Decision.

(1093) As discussed in detail in Section 13, the commitments entered into by the Notifying Party significantly reduce or even remove the overlap brought by the Transaction in the upstream markets for the production and supply of flat carbon steel products. The Commitments are also sufficient to address the Commission’s horizontal non- coordinated and coordinated competition concerns in those markets. The  Commission hence considers that the commitments would in any event remove any merger-specific ability for the merged entity to engage in input foreclosure.

(1094) As to customer foreclosure, the Commission considers that the considerations applicable to customer foreclosure in the markets discussed in Sections 9.14 and 9.15 also apply to the other vertical links. In particular, the Commission observes that the Parties are by no means important customers of flat carbon steel.

(1095) Consequently, the Commission considers that the Transaction  does  not  raise  concerns in relation to customer foreclosure.

9.16.2. Conclusion on vertical links between the production and supply of finished flat carbon steel and ArcelorMittal’s activities in other downstream markets

(1096) The Commission finds that, in view of the final commitments entered into by the Notifying Party, it is not necessary to conclude as to whether the Transaction would lead to a significant impediment to effective competition in the internal market due to input foreclosure with regard to the vertical links between the production and supply of finished flat carbon steel products, and ArcelorMittal’s activities in other downstream markets, as the commitments would in any event address such potential concerns.

(1097) The Commission further considers that the Transaction would not lead  to  a  significant impediment to effective competition in the internal market due to customer foreclosure with regard to the vertical link between the production and supply of finished flat carbon steel products, and ArcelorMittal’s activities in other downstream markets.

10. COMPETITIVE ASSESSMENT: HORIZONTAL COORDINATED EFFECTS IN THE PRODUCTION AND SUPPLY OF FLAT CARBON STEEL PRODUCTS

10.1. Legal framework

(1098) In some markets the structure may be such that firms would consider it possible, economically rational, and hence preferable, to adopt on a sustainable basis a course of action on the market aimed at selling at increased prices through a coordination of their behaviour. (714) A merger may increase the likelihood that firms are able to coordinate their behaviour and raise prices, even without entering into an agreement or resorting to a concerted practice within the meaning of Article 101 TFEU. A merger may also make coordination easier, more stable or more effective for firms that were already coordinating before the merger, either by making the coordination more robust or by permitting firms to coordinate on even higher prices. (715)

(1099) Coordination may take various forms. In some markets, the most likely form of coordination may involve keeping prices above the competitive level. Coordinating firms may also limit production or capacity, divide the market, for instance by geographic area or other customer characteristics, or allocate contracts in bidding markets. (716)

(1100)  Coordination is more likely to emerge in markets where it is relatively simple to  reach a common understanding on the terms of coordination. In addition, three conditions are necessary for coordination to be sustainable. First, the coordinating firms must be able to monitor to a sufficient degree whether the terms of  coordination are being adhered to. Second, discipline requires that there is some form of credible deterrent mechanism that can be activated if deviation is detected. Third, the reactions of outsiders, such as current and future competitors not participating in the coordination, as well as customers, should not be able to jeopardise the results expected from the coordination. (717)  In applying those criteria, it is necessary to  avoid a mechanical approach involving the separate verification of each of those criteria taken in isolation, while taking no account of the overall economic mechanism of a hypothetical tacit coordination. (718)

(1101) In examining the possibility and sustainability of coordination, the  Commission should specifically consider the changes that the Transaction brings about. The reduction in the number of firms in a market may, in itself, be a factor that facilitates coordination. However, a merger may also increase the likelihood or significance of coordinated effects in other ways. For instance, a merger may involve a ‘maverick’ firm that has a history of preventing or disrupting coordination, for example by failing to follow price increases by its competitors, or has characteristics that gives it an incentive to favour different strategic choices than its coordinating competitors would prefer. If the merged firm were to adopt strategies similar to those of other competitors, the remaining firms would find it easier to coordinate, and the merger would increase the likelihood, stability or effectiveness of coordination. (719)

10.2. Introduction

(1102) In its decision pursuant to Article 6(1)(c), the Commission considered that the Transaction raised serious doubts as to its compatibility with the internal market due to horizontal coordinated effects in the EEA markets for the production and supply of flat carbon steel. Following the Phase II investigation, the Commission's Statement  of Objections reached the preliminary view that the Transaction would result in a significant impediment to effective competition due to horizontal coordinated effects in the EEA markets for the production and supply of flat carbon steel. The Commission's objections were based on the following elements.

(1103) ArcelorMittal is the price leader for flat carbon steel across the EEA markets. It assumes the responsibility of setting pricing trends in anticipation that other main European suppliers will align their own pricing. To that effect, ArcelorMittal signals its intended pricing policy to the market, typically by means of price increase announcement in specialised press. Following such an announcement, ArcelorMittal carefully monitors whether other suppliers follow the coordinated price trend, and factors their observed reactions into deliberations for any future price increases in a repeated exercise.

(1104) In case ArcelorMittal detects that a coordinating firm deviates, it can  resort  to punitive action to discipline the rival and render the deviation unprofitable.  Typically, this would take a form of targeted selective quotes undercutting the rivals' offer to key customers of the deviating rival. By doing this, ArcelorMittal is however careful that the price effects of such actions are limited and do not trigger broader price competition.

(1105) Price coordination between ArcelorMittal and some of its European rivals may be further strengthened by an implicit market allocation arrangement, whereby such producers may tend to abstain from aggressive price competition for volumes in the natural home markets of their rivals. However, considering that the evidence suggesting such conduct (720)  mainly concerned areas  in northern Europe where  Ilva only has limited commercial presence and are therefore of limited relevance for the competitive assessment, this Decision does not further assess this issue.

(1106) The evidence suggests that while the ability of the European producers to tacitly coordinate prices was temporarily disrupted by the disturbances from the dumped import, they have followed to a significant extent a series of ArcelorMittal's price increases since 2016 which led to notable increases of prices, but also margins of European suppliers.

(1107) The Transaction would further strengthen such price coordination, notably by  removing Ilva as an independent player whose commercial policy often disrupted price stability in Southern Europe, and to a lesser degree also in Northern Europe. In addition, new structural links between ArcelorMittal and Marcegaglia, competitors  on the downstream markets for CR and GS, would likely soften competitive pressure from Marcegaglia, which sometimes aligned its policy on Ilva's.

(1108) For the reasons set out in the following Sections, and taking into account the  Notifying Party's Reply to the SO, the Commission considers that the Transaction may significantly impede effective competition as a result of horizontal coordinated effects in the EEA markets for HR, CR and GS.

10.3. The Notifying Party's arguments

(1109) In the Form CO721, the reply to the decision pursuant to Article 6(1)(c) and the Reply  to the SO, the Notifying Party contends that concerns related to coordinated effects should be readily excluded. (722)

(1110) First, market conditions are not conducive to tacit coordination. According to the Notifying Party, the European flat carbon steel industry is fragmented, with many EEA and non-EEA producers having asymmetrical market positions and conflicting incentives. The price leadership-price followership behaviour assessed in the SO is the result of unilateral competition, not coordination.

(1111) Second, the European steel markets are not sufficiently transparent to achieve and monitor coordination. According to the Notifying Party, the market is not  transparent, because demand is atomised, heterogeneous, volatile, and dispersed across countless pricing points, which are confidentially negotiated on a bilateral basis. Public price indices only offer directional insights on certain price elements of a limited number of products and, most importantly, they do not provide supplier specific information.

(1112) Third, according to the Notifying Party, the SO fails to demonstrate the existence of  an effective retaliation mechanism and does not point to any meaningful examples of retaliation.

(1113)  Fourth, constraints from EEA and non-EEA rivals are evidently enough to disrupt   any coordinated position. According to the Notifying Party, the switching of volumes between different European suppliers is incompatible with tacit coordination. Moreover, the alleged continued increase of importers' market shares even after the imposition of trade defence duties shows that imports have a disruptive effect on attempts to coordinate.

(1114) Fifth, the Transaction will not strengthen coordinated effects. According to the Notifying Party, concentration levels will not significantly increase. Ilva's price policy is not due to an intentionally disruptive strategy or a volume over price strategy, but rather due to the low quality of its products, and is not sustainable. Likewise, the Transaction would not affect Marcegaglia’s ability and incentives to compete due to the limited size of the cross-shareholdings and inability to exchange sensitive information.

10.4. The EEA markets for flat carbon steel have been prone to coordination in the past

(1115) In investigating coordinated effects, the Commission takes into account all available relevant information on the characteristics of the markets concerned, including both structural features and the past behaviour of firms. Evidence of past coordination is important  if  the  relevant  market  characteristics  have  not  changed appreciably. Likewise, evidence of coordination in similar markets may be useful information. (723)

(1116) The Commission's investigation of the Transaction revealed evidence  of  past  or recent tacit coordination  between EEA suppliers  of flat carbon steel in  a number of EEA markets, consisting essentially in coordination of general price levels as expressed in base, or effective prices to customers, in particular for HR and GS, but also other flat carbon steel products. As will be explained in more detail further below in Section 10.5.2, the coordination followed the price leader-price follower model, with ArcelorMittal taking the lead and signalling its future prices to the market in anticipation that other EEA producers will align their pricing policy. The investigation has also shown that, where the tacit coordination has in the past not been sufficiently stable, the Transaction is expected to ‘stabilise’ the market to the benefit of price coordination.

(1117) Beyond the flat steel sector, the European steel industry has in  the past  seen  a  number of cartels which were subject to a Commission decision finding an infringement of Article 101 TFEU:

(1)                            extras for steel alloys (Commission decision in 1998), involving ThyssenKrupp, Usinor (now ArcelorMittal), and others (724);

(2)                            seamless steel tubes (Commission decision in 1998), involving Ilva, Usinor (now part of ArcelorMittal), Mannesmannröhrenwerk (part of Salzgitter), British Steel, Corus (now Tata Steel), and others) (725);

(3)                            prestressing steel (Commission decision in 2010), involving ArcelorMittal, Rautaruukki (now part of SSAB/Ruukki), VoestAlpine, and others (726).

(1118) Therefore, all the major large EEA flat steel producers, notably ArcelorMittal, ThyssenKrupp, Tata Steel, Salzgitter, VoestAlpine and SSAB/Ruukki have in the past, directly or through their affiliates or legal predecessors, been found to have participated in illegal explicit coordination, albeit for products other than flat carbon steel.

(1119) Finally, the Commission takes note of the ongoing investigations by national competition authorities on a suspicion that certain producers, including  ArcelorMittal, may have colluded to fix prices of flat carbon steel in the EEA. (727)

(1120) In the light of paragraph 43 of the HMG, as elements of context, the past cartels in related steel markets show that steel producers, which to a significant degree  coincide with flat carbon steel producers, have economic incentives to seek coordinated outcomes, whether through cartels or through tacit coordination.

(1121) Against the foregoing, the Commission will subsequently assess (i) whether it would be possible for the suppliers of flat carbon steel in the EEA to reach terms of coordination (Section 10.5), and whether such coordination would be sustainable in view of (iii) market transparency and the ability to monitor deviations (Section 10.6),(iv) the existence of mechanisms for sanctioning possible deviations from coordination (Section 10.7), and (v) the (in)ability of outsiders to  disrupt coordination (Section 10.8). Last, the Commission will examine (vi) the specific changes brought about by the Transaction that would make coordination easier, more stable or more effective (Section 10.9). The analysis will be carried out jointly for the markets for  HR, CR and GS as the  considerations made below  in Sections 10.5 to 10.9 apply to all those markets (unless specified otherwise).

(1122) The Commission's findings in this Section are without prejudice to the application of Article 101 and/or Article 102 TFEU to the investigated conduct.

10.5. Major EEA suppliers likely coordinate prices following the price leader-price follower model as facilitated by market division

(1123) In the Gencor case, the General Court has found that ‘in a market with  the  appropriate characteristics, in particular in terms of market concentration, transparency and product homogeneity, those parties are in a position to anticipate one another's behaviour and are therefore strongly encouraged to align their  conduct in the market, in particular in such a way as to maximise their joint profits by restricting production with a view to increasing prices. In such a context, each trader is aware that highly competitive action on its part designed to increase its market share (for example a price cut) would provoke identical action by the others, so that it would derive no benefit from its initiative. All the traders would thus be affected by the reduction in price levels.’ (728)

(1124) The Commission finds that the conditions of competition in the markets  for  flat carbon steel in the EEA have allowed and will continue to allow the remaining EEA steel producers post-Transaction to reach a common perception as to how coordination should work and to implement such coordination.

(1125) In particular, coordination would likely take place pursuant to a price leader-price follower coordination model. This model is characterised by the existence of a strong market leader, recognised as such by other suppliers and customers for a given product and geographic market, or a sub-segment thereof. In this form of coordination, rival suppliers share a common understanding that the market leader acts as a price leader, who signals its future pricing policy, notably by public price increase  announcements,  or  more  generally its  perception  of  the  market situation going forward, in anticipation that other suppliers adapt their pricing or commercial policies accordingly (fully or at least in part). (729)

(1126) Where the market situation is sufficiently stable, it provides suppliers with an  incentive to coordinate the general level of prices of the product at issue (in the present case, base prices and effective prices for flat carbon steel), as  they can achieve higher prices as they would by engaging in competition on price. However, any price increase initiated by the price leader will only be successful if it is sufficiently followed by rivals. Otherwise, the price increase would be unprofitable for the price leader as it would lose critical volume to rivals not following the price increase.

(1127) This concept of price coordination is similar to the coordinating behaviour described  in a sizeable economic literature, and commonly referred to as ‘Collusive Price Leadership’. (730)

(1128) Given the particular features of the price leader-price follower coordination model, symmetry between firms is not required for the purpose of coordinating pursuant to this model. For this reason, the Notifying Party' argument that the asymmetric structure of the market as such renders it not conducive to tacit coordination does not appear convincing. (731) As the aforementioned literature recalls, tacit price coordination is not limited to markets characterised by symmetry in size, information etc. between main operators, but can also occur in other settings, such as the price- leader-price follower coordination.

(1129) In the present case, as described in more detail in the following subsections, the Commission’s market investigation indicates that (1) base prices could provide a relevant  focal point  for  price leader-price follower  coordination (Section    10.5.1); (2) ArcelorMittal is a price leader in this coordination in which most other large EEA suppliers participate as price followers (Section 10.5.2), (3) ArcelorMittal would be able to actively signal its forward looking pricing policy in order to facilitate alignment by rivals (Section 10.5.3); (4) ArcelorMittal would be able to continuously monitor rivals’ reaction to inform its decisions on future pricing signals to the market (Section 10.5.4); and that (5) EEA suppliers have coordinated prices pre-Transaction at least in Northern Europe and possibly, to a more limited extent, also in Southern Europe (Section 10.5.5).

(1130) While these subsections analyse past conduct of EEA suppliers, the Commission considers that post-Transaction, the market conditions will be as conducive to price coordination, if not more, as pre-Transaction. This is further assessed in Section 10.9 on the coordinated effects specific to the Transaction.

10.5.1. Base prices provide a relevant focal point for price coordination

(1131) First, while various flat carbon steel products are differentiated notably by grade, surface quality, coating type and dimensions, as explained in Section 6, the flat carbon industry benchmarks prices for each type of flat carbon steel markets (HR, CR, GS, TP) by referring to certain commodity grades, which serve as comparator for prices offered by different suppliers.

(1132) The most common benchmark is the ‘base price’, which denotes the prices for the  basic steel grades for a product type (HR, HDG, etc.), and is calculated free of any extras concerning additional product features or services. The base price is determined by suppliers typically for the spot sales and short term supply  agreements. Market intelligence agencies, such as SBB/Platts and MEPS, continuously gauge base prices based on data from a representative sample of actual transactions gathered in periodic contacts with market participants, and aggregate them to provide periodical price indices (including daily updates). Such price indices render the prices in the spot markets transparent at an aggregate level, and indicate the overall direction of pricing trends. Moreover, the base price plus extras model  also allows EEA suppliers to gather, notably through contacts with customers, price intelligence on their rivals price quotes.

(1133) With specific regard to ArcelorMittal's prices, the Commission notes that [ArcelorMittal commercial policy] does not appreciably affect the ability of other market participants (competitors, customers and others, such as market intelligence agencies)to determine base prices applied by ArcelorMittal. [ArcelorMittal commercial policy]. (732) […] (733), it does not affect the overall transparency of the price levels it charges for certain product types, […]. (734) Thus, both effective and base prices of EEA suppliers can be captured both by price indexing agencies as well as indirectly by EEA producers' contacts with customers.

(1134) Second, the relevant price benchmarks may vary geographically. As demonstrated in Section 7.2.5.7, there is a certain degree of difference between the HR prices in southern Europe and the prices charged in northern Europe, to the extent the difference is reflected in the existence of separate price indices, respectively for southern and northern Europe. [Internal document]. (735) Thus, while following the same mechanism, price coordination could lead to different price levels in Northern Europe and Southern Europe, and possibly even further differentiated at specific country level.

(1135) Third, as base prices are often a component in the pricing of specialty products, coordination of base prices could also affect the pricing of specialty flat carbon  steel products. Moreover, as base price indexes may also be used as a component in the pricing of supply agreements, coordination on base prices goes beyond spot sales, and can also impact pricing under longer term sales arrangements.

(1136)  Fourth, it appears that price coordination could potentially also extend to tariff lists  for the ‘extras’. The lists of tariffs for the extras are public. The market investigation revealed that the tariff lists of various competitors are to a large extent aligned. Although prices for extras by EEA producers are subject to much less frequent changes than base prices, [internal document].

Figure 111 – […] (736)

[…]

(1137) Fifth, price coordination on base prices for HR, CR and GS may also extend to steel distribution. This would occur notably through steel service centres (SSCs), some of which are also operated by steel producers. ArcelorMittal operates a network of  SSCs through its distribution subsidiary, AMDS. SSCs are customers of flat carbon steel producers and account for a significant share of flat carbon steel supplies. When steel producers increase base prices for flat carbon steel, third party SSCs have an incentive to pass on the price increases and follow the lead of SSCs operated by vertically integrated steel suppliers. […]. (737)

(1138) The Commission concludes that the base price, or the effective price for the corresponding product, can act as a focal point for the coordination. Price levels for flat carbon steel are transparent. (738) If rival firms adhere, at least partially, to the pricing trend initiated by the price leader, this will be reflected in the pricing for spot sales, and a new price equilibrium will be reached. Subsequently, the new transactable spot price levels will eventually trickle down to price indices.

10.5.2. ArcelorMittal as the price leader for flat carbon steel in the EEA

(1139) ArcelorMittal is generally recognised as the price leader for flat carbon steel in the EEA which pursues a policy of price discipline. In particular, it prioritises prices and margins over turnover volumes. In this respect, a customer noted that: ‘ArcelorMittal has a very rigid pricing policy and is even willing to reduce capacity so as to maintain stability in prices.’ (739)

(1140) First, ArcelorMittal is the leader in the flat carbon steel markets in the EEA  concerning production and market share.

(1141) ArcelorMittal is larger than the combined size of the two runner ups, ThyssenKrupp and Tata, in each of HR, CR and GS. Quantitative evidence on ArcelorMittal’s leading position has been presented in the description of the market structure for  these products in Section 9.2 above. In its internal documents, ArcelorMittal itself often recognises that it occupies a leading position in the EEA. (740)

(1142) Second, ArcelorMittal is perceived as the price leader by the customers. A large majority of customers reported that steel producers regularly or occasionally align their price levels to the ones offered by another competitor. (741) ArcelorMittal is generally seen as the price leader, both on account of its market presence and its past conduct. A large majority of (industrial) customers consider that ArcelorMittal is, regularly or occasionally, taking the lead by carrying out price modifications which are followed by other EEA producers. Such customers observed, for example:‘As market leader, AM price annou[n]cements are followed by the market’;‘the[y] are the biggest supplier on the market. When they move up or down, other suppliers are bound to follow to a certain extent.’;‘because of the size of AM but also due to the consolidated steel supply base in Europe, suppliers tend to follow similar price trends’;‘As the leader in Europe, most other mills wait and follow their lead.’;‘If AM indicates an increase in pricing for the next term, other suppliers typically follow the trend’. (742)‘Already today, when we want to discuss about prices with EEA producers. we haven't any information if ARCELOR haven't announce[d] their increases. Nobody make[s] prices [offers].’‘If the announcements come from AM as market leader, has more impact than announcements from other suppliers.’ (743)

(1143) Third, ArcelorMittal's role as the price leader is corroborated by its internal  documents. These confirm that [internal doucment]:

[…] (744)

[…] (745)

[…] (746)

[…] (747)

[…] (748)

(1144) Fourth, the Commission considers that other coordinating firms (price followers) would be primarily other large EEA producers both in Northern and Southern Europe. Price coordination likely takes place in particular in the regions where ArcelorMittal has strong local presence, and would be further facilitated in Southern Europe where the Parties' activities particularly overlap. [Internal document].

10.5.3. ArcelorMittal signals future pricing policy to guide rivals

(1145) […], the Commission considers that ArcelorMittal could potentially continue its signalling  strategy post-Transaction.  Although  such  strategy would  not  be strictly necessary for the coordination mechanism to be successful, it may (if implemented) further facilitate reaching terms of coordination post-Transaction.

(1146) First, ArcelorMittal would likely have an incentive to continue signalling its pricing policy going forward as to incentivise competitors to follow, as coordination makes the price increase more certain and more profitable compared to the situation where a company attempts to influence the market price on its own. (749)

(1147)   [Internal document].

[…] (750)

[…] (751)

(1148)  Third, ArcelorMittal appears to use two main channels to communicate its current  and forward-looking pricing policy to the market, namely public announcements and […].

(1149)   […]: (752) […].

(1150) An example of a price increase published by the Metal Bulletin demonstrates how  such company information is subsequently relayed to the market:

(1151)   [Internal document]: […]. (754)

(1152)   [Internal document]: […] (755); […] (756); […] (757).

(1153)   […]: […]. (758)

10.5.4. [Internal document] (1154) [Internal document]. (1155)    […] (759). […].

(1156)   [Internal document] (760) […].761 (1157) […]: (762) […].

10.5.5. Evidence of price coordination pre-Transaction

(1158) Evidence suggests that, in the past, EEA producers have followed the price policy set by ArcelorMittal as the price leader in Northern Europe and, to a more limited  extent, possibly also in Southern Europe.

(1159)   First, [internal document]. This is shown in Figure 113 and Figure 114.

Figure 113 – […] (763)

[…]

Figure 114 – […] (764)

[…]

(1160)    […]. (765) […].

(1161)   […], as shown in Figure 115.

Figure 115 – […] (766)

[…]

(1162)    [Internal document] (767) […].

Table 65 – […] (768),   (769),  (770)

[…]

(1163)   [Internal document].

(1164)   [Internal document]. (771) […].

(1165) Second, customer testimonies confirm the finding that the EEA  producers  often follow ArcelorMittal's pricing policy. (772)

(1166) In conclusion, the above evidence points in particular to the likely existence of price coordination in Northern Europe pre-Transaction in view of significant followership. The evidence in Section 10.7 also suggests that attempts to arrive at coordinated  price levels in Southern Europe have been made despite the existence of disruptors.  In view of the indications that also producers in Southern Europe have followed price increases, price coordination in Southern Europe, even if in a less stable form or occurring more sporadically, cannot be excluded pre-Transaction.

10.6. Market transparency and ability to monitor deviations

(1167) Coordinating firms are often tempted to increase their share of the market  by  deviating from the terms of coordination, for instance by offering lower prices, secret discounts, increasing product quality or capacity or trying to win new customers. Only the credible threat of timely and sufficient retaliation keeps firms from deviating. Markets therefore require sufficient transparency to allow the coordinating firms to monitor to a sufficient degree whether other firms are deviating, and thus to know when to retaliate. Transparency in the market is often higher when the number of active participants in the market is lower. Further, the degree of transparency often depends on how market transactions take place in a particular market. (773)

(1168) The Commission takes the view that the markets for flat carbon steel in the EEA are sufficiently transparent to enable coordinating firms to monitor any deviations from the possibly coordinated price level for the following reasons.

(1169) First, the markets for flat carbon steel are concentrated and only  feature  a  few sizeable suppliers. (774)

(1170) Second, there are several price indices for commodity steel (such as MEPS, SBB/Platts), which provide timely updates on the evolution of steel base prices in  the EEA, and parts thereof (such as the SBB Southern Europe index). These indices are often used as reference price in supply agreements with steel customers. Moreover, prices of the extras are generally public in the flat carbon steel producers' websites. A number of steel customers participating in the market investigation confirmed that such indices are used as a reference for spot purchases. (775) While the indices do not provide transaction specific information and are in themselves insufficient to discern an individual company's departure from the coordinated price level, they offer directional insight into price evolution and a general inference as to whether a price increase announcement has gained traction. (776)

(1171) Third, market transparency is further increased by public announcements by steel producers ([…] (777)), either concerning their capacity (such as maintenance,   closures) or their future pricing policy, and through individual commercial negotiations with customers. [Internal document].

(1172) Fourth, the vast majority of customers who participated in the market investigation consider that the steel suppliers are aware of their rivals' commercial conditions and their customer base. (778)  The following examples provide some explanations for this:‘because of high transparency (public price announcements by steel suppliers) in the market’;‘price announcements are done in press’;‘it is not excluded that they do it because their offers are often very similar’;‘for base base price yes, because prices were tracked by MEPS, Dow Jones, Steel Business Briefing. Surcharge prices for qualities, pickling, tolerances were listed on web sites, Transport costs are known’;‘it could be that a customer provide conditions to the supplier’;‘due to limited source of HR, and QP for the energy market and the few energy pipe projects around it is not difficult for traders and producers to know and therefore communicate the supply conditions in this industry’;‘there are several reports and through discussions with customers they are aware of several information’;‘market is well known by the its actors and we guess that mills know quite well what the other mills are offering’.

(1173)   Fifth, ArcelorMittal [internal document]. (779)  […].

Figure 116 – […]

[…]

(1174)   […].

(1175) The market investigation suggests that also other EEA producers receive information on rivals offers through their contact with customers (780):

‘in some cases customer tell us the prices offered by competitors during our price negotiations’;

‘We are getting on a regular basis info from customers talking to them’.

(1176) Sixth, ArcelorMittal, as a major purchaser of flat carbon steel through its SSC distribution network, has extensive access to information on competitive offers. This is compounded by the steel procurement of its trading arm. Similarly, most other EEA producers have vertically integrated SSC networks, and certain of them, such as ThyssenKrupp and SSAB-Ruukki also operate own trading arms. (781) Overall, the market investigation showed that EEA producers purchase significant volumes of flat carbon steel from their EEA rivals (782), which further enhances the price transparency in the markets.

(1177)   Seventh, [internal document].

(1178)   In the first place, [internal document] (783), as exemplified by the following document:

Figure 117 – […] (784)

[…]

(1179)   In the second place, evidence concerning [internal document]. (1180)    […]. (785)

(1181)   […].

(1182) Eighth, as regards the Notifying Party's claims that the market is not sufficiently transparent to allow firms to reach coordination, or to monitor deviations, the Commission notes the following.

(1183) In the first place, the Notifying Party contends that the demand is atomised and heterogeneous (786), resulting in a very high number of price points, rendering coordination impossible. Moreover, it alleges that it has no systematic way of tracking competitors' prices and therefore any specific price information is only anecdotal. (787) The Commission does not contest that the customer base is fragmented. However, evidence presented in this heading shows that ArcelorMittal has, through its sales network, the ability to monitor competitors' prices to individual customers and timely detect and deviations. The increases to the base price, or the effective price for the corresponding product, can act as a focal point for the coordination. (788) ArcelorMittal will thus be able to monitor whether competitor's price quotes to customers individually follow the direction set by ArcelorMittal's publicly  announced pricing policy. It will also be able, on a more aggregate level, to understand if the price levels of quotes by a specific supplier generally undercut the coordinated price level. […], which can provide a more reliable insight if an  observed pricing conduct of a competitor is a one-off occurrence or more generally reflects its pricing policy.

(1184) In the second place, the Notifying Party claims that the price indices do not offer visibility on competitors' prices. (789) The Commission does not contest that such indices normally do not allow the price leader to detect that a specific supplier deviates from coordinated price levels. However, price indices may increase transparency on a more aggregate level. They allow ArcelorMittal to detect whether its pricing policy is generally followed by other European players. Price movements reported in indices may also signal to coordinating firms that other suppliers adhere  to a given policy of the price leader and therefore further encourage followership.

10.7. Mechanisms to deter deviations from a coordinated outcome

(1185) The ability to retaliate promptly and effectively against deviating firms is key for the sustainability of price coordination. Without the threat of retaliation, firms have an incentive to reduce prices from the artificial, coordinated price level and profitably increase volumes and market share. Therefore, coordination is not sustainable unless the consequences of deviation are sufficiently severe to convince coordinating firms that it is in their best interest to adhere to the terms of coordination. Such a threat is only credible if, where deviation by one of the firms is detected, there is sufficient certainty that some deterrent mechanism will be activated (790), and if such deterrence  is timely (791). The deterrence by way of a price war or a significant output increase may entail a short-term economic loss for the retaliating firms. However, this does not necessarily remove the incentives for retaliation since the short-term loss may be smaller than the long term benefit of retaliating resulting from the return to the regime of coordination. (792) Retaliation need not necessarily take place in the same market as where the deviation occurred. If the coordinating firms have commercial interactions in other markets, such product or geographic symmetry may offer alternative opportunities for retaliation. (793)

(1186) The Commission's market investigation confirms the existence of effective and deterrent mechanisms against firms deviating from coordinated price levels (notably to capture market share), or against expansion threatening to disrupt the competitive landscape prone to coordination. These examples indicate that retaliation is mostly practiced by targeting the rivals strategic customers engaged in spot sales (as  opposed to sales under long term supply agreements) as customers mostly multisource and can easily increase the share of one supplier at the detriment of another one. This allows for an expedient reaction capable of signalling that a deviation was detected and retaliated against, thus removing the incentives to  deviate.

(1187) The Commission found consistent and extensive evidence that ArcelorMittal, as the price leader, has recently taken or considered retaliatory measures with respect to those EEA producers whose actions were perceived as a threat to coordination. These measures notably took the form of selective undercutting targeted at the deviating firm's significant customers.

(1188)   [Internal document]: […]. (794) […]. (1189) […].

(1190)   […].

(1191)   [Internal document].

(1192)   [Internal document] (795) […]. (1193)           

(1193)   [Internal document].

(1194)   [Internal document].

(1195)   […].

(1196)   [Internal document] (796), […].

(1197) Fifth, the Commission considers that the retaliation actions […] may be explained as attempts to discipline rivals and enforce some form of coordination. […].

(1198) Sixth, as regards the Notifying Party's claim that the Commission failed to  demonstrate the existence of an effective retaliation mechanism, the Commission notes the following.

(1199) […]. (797) The Commission notes that the effectiveness of a retaliation mechanism does not depend on whether or not it is put to use on a regular, or frequent basis, […] but rather on whether it represents a credible disciplining mechanism. […]. […] suggest that ArcelorMittal had at its disposition an elaborate and effective retaliation mechanism […].

(1200) […]. (798) The Commission notes that it needs not to show that one or more retaliatory actions actually took place, but that such actions could be threatened as a credible deterrence mechanism. At any rate, the elements put forward by the Notifying Party do not allow the conclusion that retaliatory action in the aforementioned instances  did not take place. First and foremost, retaliatory action […] can be effective even if the customers eventually do not accept the ‘retaliatory’ bid. Retaliatory price quotes may in themselves signal to the deviating firm that its conduct has been detected and acted upon by the price leader. Second, […]. What matters is the punitive level of the price offered by the retaliating firm, […].

(1201) On the basis of the above considerations, the Commission concludes that there is evidence of an effective and deterrent mechanism against firms deviating from coordinated price levels thus making price coordination sustainable.

10.8. Ability of outsiders to disrupt coordination

(1202) Based on the evidence relating to the period before the Transaction, the Commission considers that neither the remaining suppliers not participating in the coordination, whether located in or outside the EEA, nor the customers would have the ability or incentive to disrupt price coordination in the EEA.

(1203) First, Ilva has in the past often failed to follow the pricing as signalled by ArcelorMittal. However, the Transaction would remove Ilva and Marcegaglia as challengers to a possible coordination between the remainder of the EEA producers including  ArcelorMittal  as  the  price  leader. This is analysed in more  detail in Section 10.9.4.

(1204) Second, domestic EEA producers, with the notable exception of Ilva, have generally not been perceived as competing aggressively with ArcelorMittal, and are  more likely to participate in the coordination rather than to disrupt it.

(1205)   [Internal document] (799) […].

(1206) Accordingly, evidence in Section 10.5.5 suggests that price coordination has been restored since early 2016. The EEA suppliers should therefore be considered as participating in the coordination as price followers and not as third party disruptors  of coordination, with the possible exception of Ilva and other Italian producers.

(1207) As regards the Notifying Party's argument that significant past switching of volumes  in the EEA is incompatible with tacit coordination (800), the Commission notes the following. To accept the Notifying Party's argument that for tacit price coordination to be proven no appreciable switching between customers should be observed is tantamount to saying that all competition needs to be eliminated. However, price- leader-price follower coordination relates to direction and general levels of prices charged by the suppliers. […]. Therefore, a certain degree of price competition can persist even if the overall level and directionality of prices is subject to coordination. Therefore, unlike collusion based on market or customer allocation, tacit price coordination does not necessarily manifest itself by absence of any appreciable switching between competitors.

(1208) Third, imports do not appear to effectively constrain the EEA producers despite a consistent series of price increases leading to a strong recovery of profitability of ArcelorMittal. The Commission considers that its findings in Section 9.4.5 are fully applicable to the question whether pressure from imports could defeat price coordination. Against this background, the following points are of particular relevance.

(1209) In the first place, following the introduction of trade defence measures, imports no longer exercise a source of price pressure comparable to the dumping years. On the contrary, they appear to adapt to the EEA prices by only applying a moderate discount to the prevailing EEA prices. [Internal document]. (801) […] (802)

(1210) In the second place, under normal market circumstances, importers are price takers. [Internal document]. (803) This is confirmed by comments from several customers, for example: ‘Most of the import tend to follow the EU producer when they offer into the EU. If EU suppliers rise the prices, then import will quickly react and lift the prices as well, granting only a limited discount with respect to the EU producer’ (804); ‘Some non-EEA producers may follow announcements by EEA HR, but this is not systematic, and obviously EEA producers may also have a look at import prices trends before announcing price increases.’ (805); ‘Normally price from imports follow the increase announcements by EEA HR producers’. (806)

(1211) In the third place, even if imports at dumping prices were able to impact the overall price levels, as appears to be the case with dumped imports, coordination was still possible at the level of ‘premium’ or spread that the domestic EEA producers would charge over import price levels for commodity products. (807)  [Internal document]. (808)

(1212) Fourth, coordinating firms can take actions to deter disruptive entry or expansion, including long-term entry by non-EEA operators. There are indications that ArcelorMittal may have engaged in activities to deter entry or expansion by  non-EEA producers, in an attempt to reach a common understanding with would-be entrants and reserve the  EEA  markets  or  certain  sub-segments  to  incumbent  EEA producers.

(1213)   [Internal document].

(1214)   […]. (809) […]. (810)

(1215)   [Internal document] (811) […]. (1216)  […]. (812) […] (813).

(1217) This instance shows that [commercial strategy] a structural long term [commercial strategy].

(1218) Based on the above considerations, the Commission concludes that neither the remaining suppliers not participating in the coordination, nor the customers would have the ability or incentive to disrupt price coordination in the EEA, which is therefore sustainable

10.9. Coordinated effects of the Transaction

(1219) In the Commission's view, the Transaction would increase the likelihood that EEA producers of flat carbon steel are able to coordinate their behaviour and raise prices, even without entering into an agreement or resorting to a concerted practice within the meaning of Article 101 TFEU. Alternatively, the Transaction may make such coordination easier, more stable or more effective for firms that were already coordinating before the Transaction, either by making the coordination more robust  or by permitting firms to coordinate on even higher prices.

(1220) The evidence available suggests that already before the Transaction, ArcelorMittal  and its large EEA rivals were, at least during certain periods, able to reach  and sustain price coordination in particular in Northern Europe, and possibly also in Southern Europe, supported by an implicit market division in each others' home market areas. Section 10.5 demonstrates that ArcelorMittal successfully led a series of price increases as ‘[…]’, which culminated in a significant increase of spreads for all EEA producers.

(1221) It however appears that, prior to this, coordination was occasionally destabilised or rendered ineffective due to market shocks, such as the combination of, on the one hand, Ilva's aggressive price policy, and, on the other hand, market distortions due to imports at dumping prices. (814) The remainder of this Section will demonstrate that the Transaction would be likely, in the light of specific changes that it would bring  about, to render price coordination in the EEA easier, more stable or more effective for firms  that  were possibly already coordinating before  the Transaction,  either  by making the coordination more robust or by permitting firms to coordinate on even higher prices. At any rate, the Transaction would also increase the likelihood that EEA producers of flat carbon steel are able to coordinate their behaviour and raise prices, irrespective of whether they were already coordinating pre-Transaction.

10.9.1. The Transaction may facilitate the coordinating firms in reaching a common understanding as to coordinated price levels

(1222)  First, the removal of a major EEA producer in itself facilitates coordination, as there  is one less significant player to coordinate with. The Transaction further increases concentration in already concentrated markets for flat carbon steel in the EEA, combining the largest and the fourth largest EEA producer in terms of HR capacity and output. Following the Transaction, the HHI index (815) level for HR would rise  from 1549 to 1978 for the EEA.

(1223) The Transaction would thus remove a sizeable EEA producer and further increase already high concentration in the markets for the production and supply of  flat carbon steel, in the EEA.

(1224) Second, the Transaction would considerably strengthen ArcelorMittal's market  position in the EEA.

(1225) In addition, due to Ilva's present capacity and also considering the additional volumes Ilva could potentially sell (if its capacity were reactivated) not only in Southern, but also in Northern Europe, the Transaction would, in turn, also strengthen the merged entity's price leadership across the EEA. This is corroborated by ArcelorMittal's internal documents, [internal documents]. (816)

10.9.2.        The Transaction may further enhance market transparency

(1226) While flat carbon steel markets are already transparent today and have allowed for a timely detection of deviating conduct even pre-Transaction, the removal of Ilva would likely further enhance the degree of market transparency for flat carbon steel in the EEA.

(1227) First, by acquiring Ilva, ArcelorMittal would further expand its sales network as its source of market intelligence. Moreover, ArcelorMittal could draw further market insight from an extended distribution network, including not only Ilva's but also Marcegaglia's SSCs (817). [Internal document]:

–       […]. (818)

(1228) Second, unlike Ilva pre-Transaction, not known for actively communicating its forward looking pricing and commercial strategy to the market, the acquired Ilva business would likely follow ArcelorMittal's strategy of […]. This would likely increase market transparency to the benefit of other operators.

10.9.3. The Transaction may enhance mechanisms to deter deviations

(1229) By adding Ilva to its network of plants, ArcelorMittal could enhance its potential for retaliation, and deterrence against entry or expansion, making the coordination even more stable. This would be due to additional capacity at ArcelorMittal's disposal.   In particular, EEA suppliers would be facing the threat that ArcelorMittal's significantly increased available capacities are used to discipline any deviations from  coordination.

10.9.4. The Transaction is likely to appreciably reduce the scope for any third parties to disrupt coordination

(1230) The Commission considers that, due to the removal of Ilva, strategic links to Marcegaglia and an enlarged network of ArcelorMittal's plants, the Transaction would be likely to appreciably reduce the scope for coordination to be disrupted by outsiders and increase stability of coordination in the EEA.

(1231) As a preliminary remark, the Commission recalls that it does not assess the scope for disruptive competition against the economic conditions prevailing in the period of imports at dumping prices. These imports were found to be distorting the markets for steel in the EEA and were subsequently subject to trade defence measures by the EU. The Commission's assessment therefore takes into account the fact that the competitive pressure from imports has been significantly reduced compared to the period before the imposition of trade defence measures.

(1232) First, by removing Ilva as an independent competitor, the Transaction  would  eliminate a potential disruptor to a possible price coordination. Indeed, the evidence available suggests that, on several past occasions, Ilva did not comply with ArcelorMittal's pricing signals and disrupted its efforts to increase prices.

(1233) In the first place, [internal document] (819). Another document found that [internal document] (820).

(1234)   A further ArcelorMittal document describes Ilva's conduct as follows: […]. (821)

(1235)   [Internal document]. (822) […]. (823)

(1236) Moreover, as the largest incumbent in Italy, Ilva's failure to follow ArcelorMittal  likely also impacted the other Italian flat carbon steel suppliers, Marcegaglia and Arvedi. Namely, these two companies are locally exposed to Ilva and risked losing volumes if their prices were appreciably higher than Ilva's.

(1237) In the second place, [internal document] are consistent with the assessment of the Transaction by third parties. […]. (824) An industrial customer observed, in the context of the question whether ArcelorMittal is the market leader, that this was ‘Never [the case] in the past, when Ilva was the trend-setter.’ (825)

(1238) In keeping with this, several industry operators indicated that the Transaction would improve price stability (826), as exemplified by the following statement: ‘SPL expects

ArcelorMittal and Ilva taking a leading role with the effect of price stability and continuity.’

(1239)   […]. (827)

(1240) The disappearance of Ilva as a disruptor could therefore likely facilitate a possible price coordination of the remaining competitors under ArcelorMittal's price leadership.

(1241) Second, the Transaction could, by strengthening economic links between the merged entity and Marcegaglia, further reduce Marcegaglia's ability and incentives to compete independently of the merged entity. […]. (828) This is for the following reason:

(1242)   In the first place, [internal document]. (829)

(1243) In the second place, the Transaction would likely reduce Marcegaglia's incentives to compete by establishing a cross-minority shareholding between the merged entity  and Marcegaglia. [Commercial information]. (830) Notwithstanding that the cross- shareholding only accounts for […] of the respective companies' total shares, it could further align the companies' commercial and financial interests. [Internal document]. (831) Moreover, such a corporate design would establish a permanent and mutual structural link between the merged entity and Marcegaglia and could allow for exchanges of sensitive information and thus further facilitate coordination. (832)

(1244)   […]. (833)

(1245) In the third place, the Transaction may reduce Marcegaglia's ability and incentive to compete with ArcelorMittal due to an accompanying long term arrangement, by which the merged entity would supply to Marcegaglia very significant volumes of HR. Considering that Marcegaglia is a re-roller, and thus does not produce primary steel (slabs) or HR, the latter is a key input for Marcegaglia's business. The Transaction would (i) further entrench Marcegaglia's commercial dependence on supplies from the merged entity and (ii) bring about significant cost commonalities between them for downstream flat steel products as well as further transparency on cost of input, and thus facilitate price coordination on downstream products.

(1246) In the fourth place, a number of customers expressed the view that the Transaction would negatively impact Marcegaglia's commercial independence from the merged entity: (834) ‘Marcegaglia may align its commercial strategy on ArcelorMittal and consequently Marcegaglia may increase it prices’;‘they are in the same boat and marcegalia will finally have the same conditions like arcelor’;‘one player less, the remain [sic] will better control the demand and offer’; they are creating the monopol on the market, which will cause higher prices’;‘being part of the deal, even though with a smaller share, they will be even more influenced by Arcelormittal policy’;‘Arcelor marks the price to Marcegalia’.

10.10. Conclusion as regards horizontal coordinated effects for flat carbon steel products

(1247) The Commission takes the view that the Transaction may lead to a significant impediment of effective competition in the internal market due to horizontal coordinated effects in the markets for flat carbon steel (HR, CR and GS) in the EEA. However, it is not necessary to conclude on this issue in view of the final commitments proposed by the Notifying Party, which would in any event address such potential horizontal coordinated effects.

11. EVIDENCE ON THE EFFECTS OF THE TRANSACTION

(1248) The conclusions based on the Commission's assessment that the Transaction would lead to anti-competitive non-coordinated effects in the markets for HR, CR, and GS in the EEA (in particular in Southern Europe and Italy) […] which anticipates that  the Transaction would enable it to sustain higher prices (Section 11.1) further corroborated both the views of customers and other third parties (Section 11.2).

11.1. [Internal document]

11.1.1. [Internal document]

(1249)   […].

(1250)   […] (835), […].

Figure 118 – […] (836)

[…]

(1251)   […].

Figure 119 – […] (837)

[…]

(1252)   […].

(1253)   […].

(1254)   […].

(1255)   […] (838) […].

(1256)   [Internal document].

(1257)   […].

(1258)   […].

(1259)   [Internal document], (839) […]. (1260) […] (840): […].

(1261)   The Commission makes the following observations.

(1262)   In the first place, […]. Consumer harm would be likely much higher also   taking into account that higher prices would also be charged by other EEA producers.

(1263)   In the second place, […]. (1264)    In the third place, […].

11.1.2.        [Internal document]

(1265) The Notifying Party refers to a number of synergy documents which allegedly  replaced and superseded the earlier analyses and rendered them irrelevant. (841) However, it appears that these later synergy analyses were not prepared in order to inform ArcelorMittal's internal decision-making as to whether or not to acquire Ilva. [Internal document]. (842)

(1266)   [Internal document].

(1267)   [Internal document].

(1268)   [Internal document] (843) […]. (1269)              […]:

(1)     […]. (844)

(2)     […]; (845)

(3)     […] (846)

(4)     […] (847)

(5)     […]. (848)

(6)     […]. (849)

(1270)   […]. (850)

(1271)   […].

11.1.3. [Internal document]

(1272) Other documents by ArcelorMittal or in its possession corroborate the above findings [internal document], were among principal objectives that ArcelorMittal has been pursuing with the proposed Transaction.

(1273)   First, […], (851) […].

(1274)   Second, […].

11.1.4. Notifying Party's arguments claiming [Parties' submissions] are unconvincing

(1275) ArcelorMittal claims, first, that when considering a potential transaction, it typically analyses a wide array of possible scenarios, including those that are less likely to inform its decision making process. Second, according to ArcelorMittal, […]. Third, according to ArcelorMittal, the estimates only amounted to ‘back-of-the-envelope’ calculations which were later anyway dropped entirely from ArcelorMittal's internal synergy calculations.

(1276) First, the Commission notes that […]. This was confirmed by the bidding process itself, in which the AcciaItalia consortium was a serious bidder whose final bidding score was not significantly lower than ArcelorMittal's. (852)

(1277) Second, the argument that the price effects from the risk assessment would not  emanate from a competitive bidder for Ilva but due to a market distortion from State aid to Ilva is wholly unsupported by […].

(1278)   Third, […].

11.2. Customers and third parties consider that the proposed Transaction would increase the market power of ArcelorMittal and lead to higher prices

(1279) A majority of customers which expressed a view on the subject in reply to the Commission's questionnaires consider that the proposed Transaction would lead to price increases for flat carbon steel in the EEA.

(1280) First, a large majority of direct  customers of steel producers (industrial  customers  and SSCs), expect a price increase for commodity flat carbon steel (comprising both HR and the downstream products). (853) A clear minority expected no  changes  to prices, while nobody expected them to decrease. Customers commented, for example:‘since with this transition AM will have the vast majority of spot market capacity, prices will likely controlled by AM’;‘ILVA will increase the price list on Commodities. We can estimate that it could be around 5%’;‘AM will increase capacity but will also have a bigger market share and more power to impose price increases’ (854);‘very important impact as ilva was a major player for commodity’. (855)

(1281) Second, a majority of industrial customers considered that  the Transaction  would  lead to higher prices for HR. (856) ‘by acquiring Ilva's potential, AM will become main dominating European producer and will be able to dictate market conditions’ (857);‘AM will impose their price on the market’;‘AM has the Leadership on Prices of the market in all EU countries’;

(1282) The majority was even higher amongst independent SSCs which were in particular concerned with the price increase. (858) ‘especially in the nearby market 'Ilva' will adopt the AM price level In general Ilva is better priced than AM’; ‘given that there is also Anti-Dumping [duties] on hot-rolled, and now 2 strong hot- rolled suppliers are merging, [a price increase] is likely’ (859); ‘in a situation of a quasi monopoly, the price will be influenced by the new entity’; (860) ‘from the mid nineties on, Ilva had a moderating function on the Italian market to the benefit of final users and their exports. However, Arcelor's acquisition of Ilva provokes the thought of future price increases in Italy…’ (861); ‘the price will increase mainly for the customers that used to buy from Ilva at lower market prices’.

(1283) Third, a majority of customers considered that the Transaction would lead to higher prices for CR. Again, the concerns were even more pronounced amongst the customers operating SSCs. By and large the customers provided the same or similar reasoning for their view as they did for HR. (862)

(1284) Fourth, a majority of customers considered that the proposed Transaction would lead  to higher prices for GS (EG and HDG), and the SSCs were particularly concerned. (863) Customers, for example, commented: ‘by adding this factor to the anti-dumping measures, it seems clear that this will lead to a price increase trend’; ‘the increase or decrease of price on this product depends of AD decision by EEA. If AM buys Ilva, normally the price of all hot dip coated steels should increase but this market can really be impacted by imports’;‘Same as for HR. There will be less material going from Italy to Northern Europe, thus reducing the availability and enabling higher prices for steel on average in Northern Europe’; (864)‘In general, for zinc, Ilva is better priced than AM- the merger means that AM has one competitor less for EG’;‘we think that it will increase because of AM extra price list is higher than Italian extra price list’. (865)

(1285) Fifth, an overwhelming majority of customers which procure flat  carbon  steel  through SSCs consider that the proposed Transaction will result in price increases for flat carbon steel (866)and some considered that their company would be negatively affected (867). Customers, for example, commented:‘AM was always more expensive than Ilva’ (868);‘Ilva is the largest European steel mill. If ArcelorMittal acquires it, it will have a dominant position on the European market allowing it to apply new pricing terms.’ (869);‘this company will have a position of a quasi monopoly, we can fear a price increase’ (870);‘it could later lead to a generalised price increase for steel in the Italian market, also because already today AreclorMittal's prices in Europe are indicative for the entire sector’ (871);‘by AM aquiring Ilva this reduces competition in italy, leaving AM to do control the italian market to a large degree, there are other competitors in europe [b]ut none so well placed and located’;‘Having the monopol, they will for sure increase the prices immediately, as they usually did during all previous acquisition on the other markets and European countries. At the moment, in Italy the price is €50,00/ton lower than other European markets, controlled by AM.’‘Ilva was doing the best price offer. Arcelor had to review their offer according Ilva / Marcegaglia / ... . This lever is deleted.’

(1286) Customers of SSC also described how the proposed Transaction would impact their company: (872) ‘loss of our foreign markets, with their price increase we will be no longer able to supply our distributors abroad’ (873); ‘The price of the metal will increase, and we could not increase the price to our customer. We are a company working for big company. What we can do, between ARLECOR (supplier), and XXXXX (big customer)? There have already some agreements bewten this companies. To have ILVA like supplier, it was good, because we could negociate the prices. In the futur, it will be only ALRCELOR/ILVA and our customer who will decide the price’; ‘ WE WILL DIE AND CLOSE!!! There will not be convenient prices in the future, to continue our production for longer time.’;‘We work directly with ArcelorMittal and Ilva. Ilva had an attractive price positioning, which allowed us to maintain realistic price levels on the industrial market in France. If ArcelorMittal acquires Ilva, its dominant position on the French market will allow it to impose new pricing terms. We also buy material from other SSCs which source from Ilva or Arcelor. They will be equally impacted and will pass on the price increases to us. Thus, the contraction of number of supplier on the French market will impact us.’ (874);‘loss of a supplier (Ilva) which was reliable and competitive, almost no other  possible supplier than Arcelor. Price increase, because Areclor can make prices as it wants, longer lead times (as Arcelor is much less reactive than Ilva).’ (875)

(1287) The minutes of various calls between the case team and customers of the Parties suggest the view that ArcelorMittal is a price leader in the flat carbon steel industry and that the Transaction would result in less competition and higher prices in the flat carbon steel markets. Some examples are provided below: '[…] puisque Ilva n'a pas beaucoup de gamme, Ilva a une stratégie d'aggressivité de prix […] L'existence d'Ilva a une importance pour le prix de base'. (876) 'Le rachat d'Ilva par ArcelorMittal va diminuer la concurrence de façon significative […] Ilva il y avoir une hausse de prix. Beaucoup de fournisseurs n'auront plus cette concurrence'. (877) '[…] takes the view that a potential acquisition of Ilva by the ArcelorMittal and MArcegaglia consortium could have an impact on the price of steel, as it would increase the market power of ArcelorMittal. In addition, If Ilva ceases to act as a price constraint in southern Europe, there might be a convergence of prices between northern and southern Europe'. (878) '[…] does not expect the transaction to have a direct impact on its business since it currently does not have any business relations with Ilva. However, […] fears that the reduction of suppliers and the consodlidation in the industry might push market to a monopole situation'. (879) As a result of the transaction, […] expects that ILva's prices will be aligned to those of ArcelorMittal'. (880)

(1288) Moreover, the minutes of various other calls between the case team and market participants or customers suggest that ArcelorMittal is a price leader and that the Transaction would result in an alignment between prices in the south and north of Europe. Some examples are provided below:

'In a post transaction scenario, competitiveness would be significantly reduced in all of the EEA steel market. Currently, pricing levels in the North of Europe,  where ArcelorMittal is active, are slightly higher compared to the South of Europe. Competition trends also appear to differ between these two regions since the distributors in the South of Europe tend to be more independent and  less constrained, especially due to the geographic location of Italy and Spain which favour inflows of imports. Specifically, the acquisition of Ilva by ArcelorMittal will remove one important steel operator which – together with other producers – contributes to preserve a certain degree of competitiveness in the South of Europe. The current level of competition in Southern Europe allows customers to source supplies for lower prices in this region compared to Northern Europe. Following the acquisition, ArcelorMittal will control Ilva's pricing policy and the Southern European players (such as Marcegaglia, Gabrielli etc.) will have to negotiate their sourcing prices directly with ArcelorMittal. This could ultimately lead to an alignment of the pricing levels of the North and the South, as Ilva would no longer counteract ArcelorMittal’s price increases as a competitor. As a general consideration, due the price increase registered in the European steel market, it is becoming more and more difficult also for EEA companies active in the automotive industry to compete, within the EEA, with their non-EEA based    global competitors. This applies to the Company and its European customers as well.' (881) '[…] can only notice that prices are typically lower in southern Europe, partly because of the existence of independent low-cost suppliers like Ilva. In Europe ArcelorMittal is currently a price leader.[…] takes the view that if the consortium including Jindal wins the bid, it might mean that the development of ILVA will not be financed at the expense of EEA customers such as […].' (882)

(1289) Finally, in 2017 ArcelorMittal was […] contacting the  customers  potentially  impacted by the Ilva acquisition. During those contacts also the Commission's questionnaires were discussed, (883) [internal document].

(1)     […];

(2)     […];

(3)     […].

(4)     […]. (884)

(5)     […]. (885)

12. CONCLUSION ON THE COMPATIBILITY OF THE NOTIFIED CONCENTRATION  WITH THE INTERNAL MARKET

(1290) For the reasons set out in Section 9 to 11, the Commission finds that  the  concentration, as notified, would significantly impede effective competition in a substantial part of the internal market within the meaning of Article 2 of the Merger Regulation at least as a result of horizontal non-coordinated effects as regards markets for HR, CR and GS in the EEA.

13. COMMITMENTS

13.1. Introduction

(1291) In order to remove the serious doubts identified by the Commission in its Phase I investigation, the Notifying Party submitted commitments on 19 October 2017 pursuant to Article 6(2) of the Merger Regulation.

(1292) In its decision pursuant to Article 6(1)(c) of the Merger Regulation, the Commission found that the commitments were manifestly not sufficient to eliminate the serious doubts identified by the Commission at that stage of the proceedings, and  accordingly were not market tested. In particular:

(1)   The scope of the commitments did not cover all areas where serious doubts were identified;

(2)   The remedies were not clear-cut in addressing the serious doubts raised by the Commission even for those markets which were covered by the scope of the remedy, as the divestiture covered a very limited part of the overlap created by the transaction;

(3)   The remedy did not reproduce the integrated model of ArcelorMittal and Ilva also as regards their access to own distribution.

(4)   Despite the removal of structural links, it was not clear that Marcegaglia would be independent on the merged entity for the sourcing of hot rolled coils.

(1293) During the Phase II investigation and following the notification of the SO, the Notifying Party's reply to the SO and the oral hearing, the Notifying Party submitted further commitments on 15 March 2018 (the ‘Commitments of 15 March 2018’).

(1294) In order to address a number of issues that emerged from the market test and the Commission's investigation of the proposal, the Notifying Party submitted a final set of commitments on 11 April 2018 (the ‘Final Commitments’).

13.2. General principles for the assessment of commitments

(1295)  As set out in the Remedies Notice (886), the following principles apply where parties to a merger choose to offer commitments.

(1296) Where a concentration raises competition concerns in that it could  significantly impede effective competition, the parties may seek to modify the concentration in order to resolve the competition concerns and thereby gain clearance of their merger (887).

(1297) 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 (888).

(1298)  To that end, the commitments have to eliminate the competition concerns entirely   and have to be comprehensive and effective from all points of view (889). 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 (890).

(1299) Moreover, commitments must be capable of being implemented effectively within a short period of time891. 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. (892)

(1300) 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. (893)

(1301) 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. (894)

(1302) 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. (895)

(1303) Normally, a viable business is a business that 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. (896)

(1304)  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. (897)

13.3. The Commitments of 15 March 2018

13.3.1. Description of the remedy proposal

(1305) The Notifying Party formally submitted commitments on 15 March 2018 (‘Commitments of 15 March 2018’), which consisted of the divestiture of several plants   (the   'Divestment   Businesses   as   proposed   in   the   Commitments   of   15 March 2018') to an independent purchaser or purchasers subject to approval by  the Commission as well as a commitment to sever structural links with Marcegaglia. More specifically, the Divestment Businesses are structured in three packages. Each of these three packages individually has to be sold as a whole to a single buyer:

(1306) Package 1 (the ‘Piombino/Galati Business’) comprises ArcelorMittal's  flat  carbon steel assets in Piombino (Italy, the ‘Piombino Plant’), Galati integrated steelworks (Romania, the ‘Galati Plant’) as well as a plant in Skopje (Former  Yugoslav  Republic of Macedonia, the ‘Skopje Plant’) which is a satellite plant to the Galati Plant. The package includes operational capacity to produce crude steel, HR, CR, HDG, OC and QP. In addition, a limestone mine, a stake in a sea terminal and a port on a connecting river close to the Galati Plant are included, as well as a stake in a transportation company operating in the port of Piombino.

(1307)   Moreover, ArcelorMittal committed [commercial information].

(1308) Package 2 (the ‘Liège/Dudelange Business’) comprises certain ArcelorMittal's flat carbon steel assets in Liège (Belgium, the ‘Liège Plant’) and Dudelange (Luxembourg, the ‘Dudelange plant’) and includes the operational capacity to produce CR, HDG, EG and TP. It furthermore includes a steel service centre, and an option to acquire a batch annealing shop.

(1309)   [Commercial information].

(1310)   Furthermore, the Notifying Party committed to [commercial information].

(1311) Package 3 (the ‘Ostrava Business’) comprises ArcelorMittal's flat carbon steel assets  in Ostrava (Czech Republic, the ‘Ostrava Plant’) which includes the operational capacity to produce crude steel, HR and long products. An engineering workshop  and a tooling services company are also included. The business furthermore includes distribution assets in Italy.

(1312)   Table 66  provides  an  overview  of the  total  operational  capacity contained  in the Divestment Businesses.

(1313) The Commitments of 15 March 2018 therefore included six plants. In addition, the Commitments of 15 March 2018 provided that certain entities and assets be explicitly excluded from the Divestment Business. Those exclusions relate mainly to:

–       […].

(1314) Additional assets are included to ensure the viability and competitiveness of the Divestment Businesses. These are mainly:

–                   Governmental authorizations, including environmental permits

–                   Service, use, lease, utilities supply, and other agreements, contracts and commitments

–                   Copies of books of accounts, customer lists, order books, and non- commercially sensitive correspondence, records, and files

–                   Sales contracts and order book

–                   Personnel

(1315) The Notifying Party also committed that it shall cause Marcegaglia to divest its shareholdings in the AM Consortium. ArcelorMittal shall also disengage from its obligation to acquire a shareholding in Marcegaglia, […].

(1316)   Finally, […].

13.3.2. The Notifying Party's views

(1317) According to the Notifying Party, the Divestment Business(es) as proposed in the Commitments of 15 March 2018 would be self-sufficient, effective and viable competitor(s) to the Parties:

(1318) With regard to the production and supply of GS products, the Divestment Businesses' current operational GS capacity matches that of  Ilva’s with respect to both HDG   and EG. According to the Notifying Party, the assets are also capable of producing  all grades and types of products that Ilva is.

(1319) As to the production and supply of HR and CR, the Divestment Businesses include  two highly competitive HR assets which have undergone significant capex investments recently. These replicate Ilva's HR/CR commodity and specialty product capabilities and, according to the Notifying Party, the available HR/CR capacity of these businesses significantly exceeds Ilva's HR/CR sales.

(1320) The Divestment Businesses have the logistic capabilities including  through  established rail, road and maritime logistic networks to enable them to profitably supply GS throughout the EEA, including in Southern Europe. Moreover, the transport costs of delivering to Southern Europe are claimed to be comparable to those of Ilva's site in Taranto.

(1321) The divestment of the Divestment Businesses would result in ArcelorMittal's market share to drop to below 35% in the EEA under any metric (sales share, capacity share or consumption share), for HR, CR, HDG and EG. Furthermore, this would increase the capacity in the hands of rivals.

(1322) Furthermore, in the Notifying Party's view, the plants included in the Divestment Business as proposed in the Commitments of 15 March 2018 constitute financially strong and stable assets. The Divestment Businesses are commercially viable, generate significant revenues and positive EBITDA, are standalone and integrated and attract significant interest from potential buyers. Taken as a whole, the Divestment Businesses as proposed in the Commitments of 15 March 2018 are financially sound and it could be expected that several high quality bidders would compete to acquire these businesses.

(1323) According to the Notifying Party, the Commitments of 15 March 2018 can be implemented effectively within a short period of time. The Divestment Businesses are physically separate from other ArcelorMittal assets and there are no legal or regulatory obstacles to sale.

(1324)   […]. (900)

(1325) Concluding on their suitability to remove the significant impediment to effective competition to which the Transaction would otherwise give rise, the Notifying Party claims that the Commitments of 15 March 2018 would address all the concerns identified by the Commission. The Divestment Businesses will be operated as self- standing integrated businesses, capable of profitably supplying the EEA, including Southern Europe and with product capabilities replicating Ilva's constraint. Furthermore, the Marcegaglia Commitment would ensure the removal of all  structural links between the two companies that were to be created by the Transaction, eliminating any scope for concern in this regard.

13.3.3.        Responses to the market test of the Commitments of 15 March 2018

(1326)  On 15 March 2018, the Commission launched a market test of the Commitments of  15 March 2018.

(1327) The market test was intended to assess whether the divestment is suitable to remove competition concerns in the HR, CR and GS markets. In addition, the market test was intended to test whether the scale and scope of the businesses being divested, including the transfers of key personnel, and the transitional agreements and the proposed commercial conditions of those agreements were sufficient to ensure its viability and competitiveness, and whether certain carve outs and discontinued supply flows impaired such viability and competitiveness.

(1328)  The results of the market test were overall positive. A majority of the respondents   that took a position on the remedies expressed a positive opinion regarding the scope of the Commitments of 15 March 2018 and their effectiveness in removing the competition concerns.

(1329) The Commission has reviewed all individual responses and its assessment of the market test is based on the totality of the replies, with a particular focus on the replies that expressed substantiated opinion. The responses include the feedback from competitors and customers in all areas where the Commission raised competition concerns, as well as from interested third parties.

(1330) With regard to the production and supply of HR, a majority of customers that took a position on the question consider that the proposed commitments would in principle be suitable to effectively remove the Commission's competition concerns. (901) A majority of all competitors responding considered the same. (902)

(1331) With regard to the production and supply of the downstream products, a majority of customers that took a position on the question consider that the proposed commitments would in principle be suitable to effectively remove the Commission's competition concerns with regard to each of the downstream products (CR, HDG  and EG). (903)  A majority of all competitors responding considered the same. (904)

(1332) A majority of the customers that took a position consider that the scale and scope of  the business to be divested is sufficient to ensure the Divestment Business'  immediate viability and competitiveness on the market. When considering a longer horizon, even a higher share of the customers taking a position is positive on the  scale and scope of the business to be divested. (905) Competitors are markedly even more positive for both the immediate viability and competitiveness of the divested assets going forward. (906)

(1333) A crucial point of attention in the market test was the ability of the purchaser of the Divestment Businesses to secure sufficient HR supplies at economically viable conditions as an input to the CR and GS operations of Piombino/Galati and Liège/Dudelange. [Commercial information]. (907) […]. (908)

(1334)   [Commercial information]. (909)

(1335)   [Commercial information]. (910)

(1336)   [Commercial information].

(1337) Finally, while the market test replies also indicated a concern about the fact that IP rights were not included in the divestment, the Commission notes that the Notifying Party's IP rights are in relation to some specific specialty products proprietary to ArcelorMittal, rather than commodity products, which is where Parties' activities overlap. The absence of these intangible assets should therefore not affect the viability of the Divestment Business.

(1338) The exception is the Galati Plant, as certain intangible assets are required for the operation of the Plant. In this respect the Commitments of 15 March 2018 included the required forms of know-how essential for the operation of BF5 and BOF.

13.3.4.         The Commission's assessment

13.3.4.1. Structural divestiture of capacity under the Commitments of 15 March 2018

(1339) Table 67 shows the capacity shares of the Parties in the EEA taking into account the divested capacity included in the Commitments of 15 March 2018. The table is divided in two panels. The top panel shows the Parties' capacity shares absent the remedies under Scenario 1 (that is readily available capacity, see recital (491)). The bottom panel shows the capacity shares considering the impact of the remedies, and hence accounting for the capacity of the operational assets of the divestment businesses (including the reactivation of the HDG Liege line 4) in the capacity shares as computed in the top panel. (911)

(1340)   In  relation to HR, the  operational capacity of  the Divestment Business amounts    for [0-5]% of HR capacity in the EEA, and represent about [40-50]% of the overlap in capacity between the Parties […]. This figure, however, does not account [commercial information].

(1341)  In relation to CR, the capacity of the Divestment Business amounts for [5-10]% of   the  CR  capacity in  the  EEA,  and  represent  more  than  [80-90]%  of  the capacity overlap between the Parties. According to the Notifying Party, with limited investments the CR capacity could be easily increased to a level above the overlap.

(1342) In relation to GS, the capacity of the Divestment Business, taking into account only operational assets, amounts to [5-10]% of the GS capacity in the EEA. This is composed of [5-10]% of HDG capacity in the EEA and [0-5]% of EG capacity in the EEA. The capacity of HDG is slightly above the capacity overlap between  the Parties, whereas the capacity of EG is slightly below the capacity overlap between the Parties. As the HDG capacity in the EEA is significantly larger than the EG capacity, the aggregate GS capacity remains slightly higher than the GS capacity overlap between the Parties. Those figures do not take into account lines included in the divested businesses, but that are currently mothballed. Specifically, the HDG  Line 2 in Piombino, and the EG Line 2 in Dudelange.

(1343)  Table 68 and Table 69 show the notional sales shares of the divested businesses in   the EEA and Southern Europe, respectively,  based  on  ArcelorMittal's  sales  data for 2016. Regarding HR, the proposed divested businesses account for [0-5]% of sales in the EEA, which represents more than half of Ilva's current sales and the overlap. Regarding CR, the proposed divested businesses accounts for [0-5]% of sales in the EEA, which again represents more than half of Ilva's current sales. Regarding GS, and similarly for HDG and EG, the divested businesses account for [5-10]% of EEA sales, which is more than the overlap represented by Ilva's sales.

(1344) In Southern Europe, the notional sales shares of the divested businesses would be  lower than at EEA-level. Specifically, the combined sales shares of the divested businesses would  amount  to  [0-5]%  for  HR,  [0-5]%  for  CR  and  [5-10]%  in  GS ([5-10]% in HDG and [5-10]%).

(1345)  Sales shares in Southern Europe, however, reflect the current patterns of distribution  in the EEA. In this respect, ArcelorMittal's multi-plant network is likely to heavily influence the sales distribution of the Divestment Business, and, hence, their combined notional sales shares. Indeed, it is likely that, because of its network, the current sales of ArcelorMittal are concentrated around its plants. Therefore, the pattern of sales observed for ArcelorMittal is likely to misrepresent the sales patterns that the Divestment Business will have once controlled by a different entity. In particular, it is likely to underrepresent the sales of the divested businesses to Southern Europe and in particular Italy. This is because, once detached from ArcelorMittal's network, the Divestment Business' distribution patterns will be unconstrained, which may favour higher consumption areas, such as Italy. Moreover, the distribution assets included in the remedy package will provide a further  incentive for the Divestment Business to sell to Southern Europe and to Italy.

(1346)  The sales shares of the Divestment Business reported in Table 68 and Table 69 also  do not account for a number of expansions that are currently foreseen or incentivised [commercial information].

(1347) All in all, the Commission considers that sales shares, in particular for Southern Europe, underestimate the future sales of the Divestment Business and likely misrepresent their distribution pattern across the EEA. The same shortfalls would apply equally to consumption shares, which are therefore less reliable for the purpose of assessing the remedies. The suitability of the remedy to remove the significant impediment of effective competition brought about by the Transaction is better assessed considering the structural capacity to be divested.

13.3.4.2. Suitability to remove the significant impediment to effective competition in relation to horizontal non-coordinated effects in the production and supply of HR products

(1348) In assessing the suitability of the Commitments of 15 March 2018 to remove the significant impediment to effective competition in relation to production and supply of HR products, the Commission takes into account the following factors.

(1349) First, the Commitments of 15 March 2018 sufficiently reduce the overlap in HR brought about by the Transaction.

(1350) While the Divestment Business consists of three packages, HR assets are only part of Package 1 (Piombino/Galati Business) and Package 3 (Ostrava Business). More specifically, the Piombino/Galati Business includes an integrated flat carbon steel plant in Galati, Romania. [Commercial information]. (915)

(1351) The Ostrava Business includes ArcelorMittal's flat and long carbon steel production plant in Ostrava, including its four blast furnaces, […]. The HR operational capacity of the Ostrava Plant is […]. (916)

(1352)  Thus, the combined HR  operational  capacity of the Divestment  Business amounts   to [commercial information].

(1353) While the Divestment Business does not cover the full overlap as regards production  of HR, the Commission considers that: (i) the increment created by the Transaction  in HR is smaller than in GS; (ii) the level of competitive constraint that imports exert in the area where the effects of the Transaction would be most prevalent, namely Southern Europe and Italy, is stronger for HR; (iii) the scope of the HR remedy, as proposed in the Commitments of 15 March 2018, brings down the merged entity's capacity share to about [40-50]%, with a very limited increment. The distribution assets included in Package 3 should incentivise the purchaser of the Ostrava Business to sell volumes of HR into Southern Europe and in particular in Italy.

(1354) Second, as regards the product portfolio of the Divestment Business, the Commission considers that, based on information provided by the Notifying Party, the Divestment Business will have the ability to produce the HR product grades that Ilva produces. This finding is also supported by the results of the market test in that the majority of both competitors and customers taking a position suggest that the capabilities of the Divestment Businesses are at least similar to those of Ilva’s. (917)

(1355) Third, the Commission's assessment also considers the ability of the Commitments of 15 March 2018 to address the competitive concerns in the area of the market where the concerns are most prominent, namely Southern Europe and Italy. In case of HR, the relevant question is whether there would be the ability and incentives for the Divestment Business to supply HR to Southern Europe, given that the HR assets are located in Romania and the Czech Republic.

(1356)  In this respect, the inclusion in Package 1 of the finishing lines in Piombino, as well  as the inclusion in Package 3 of distributional assets in Italy, will facilitate and incentivise the remedy taker(s) to deliver HR volumes from Galati and Ostrava to Southern Europe and in particular Italy. However, Package 1 does not include any distribution assets in Southern Europe.

(1357) Moreover, while Galati and Ostrava are not located in Southern Europe, as defined in recital (339), low transport costs will allow the purchaser of the Divestment Business (in particular of Packages 1 and 3) to have the ability and incentive to supply HR from the Galati and Ostrava plants into Southern Europe, [commercial information].

(1358) In conclusion, the Commission finds that the Commitments of 15 March 2018 are sufficient to fully remove the significant impediment to effective competition in relation to production and supply of HR products to which the Transaction would give rise in the EEA. However, this is subject to improvements on technical matters in order to strengthen the incentives of the purchaser of Package 1 to serve customers in Southern Europe and in particular Italy, for instance in the form of distribution assets.

13.3.4.3. Suitability to remove the significant impediment to effective competition in relation to horizontal non-coordinated effects in the production and supply of CR products

(1359) In assessing the suitability of the Commitments of 15 March 2018 to remove the significant impediment to effective competition in relation to production and supply of CR products, the Commission takes into account the following factors.

(1360) First, the Commitments of 15 March 2018 sufficiently remove the overlap in the operational CR capacity brought about by the Transaction

(1361) While the Divestment Businesses consists of three packages, CR assets are only part  of Package 1 (Piombino/Galati Business) and Package 2 (Liège/Dudelange Business). Within Package 1, CR assets are found at all production sites included: in Piombino (Italy), Galati (Romania) and Skopje (Former Yugoslav Republic of Macedonia).

(1362) The Piombino Plant comprises ArcelorMittal’s flat carbon steel plant in Piombino (Livorno, Italy) […].

(1363) The Liège/Dudelange Business comprises certain of ArcelorMittal's flat carbon steel lines in Liège, Belgium and ArcelorMittal's flat carbon steel plant in Dudelange, Luxembourg. […].

(1364) The Liège Plant includes, inter alia, [commercial information]. It also includes [commercial information] (918).

(1365) The combined CR operational capacity of the Divestment Business  (Piombino,  Galati, Skopje and Liège) is […], which represents approximately […]% of Ilva's operational overlap in terms of capacity.

(1366) Second, as regards the product portfolio of the Divestment Businesses, the  Commission considers that, based on the information provided by the Notifying Party the Divestment Business will have the ability to produce the CR product grades currently produced by Ilva. The results of the market test support this finding in that the majority of both competitors and customers taking a position suggest that the capabilities of the Divestment Businesses in CR are similar to those of Ilva. (919)

(1367) Third, the Commission's assessment also considers the ability of the Commitments of 15 March 2018 to address the competitive concerns in the area of the market where the concerns are most prominent, namely Southern Europe and in particular Italy. In this respect the Commission notes that while only the Piombino Plant is located in Italy, its operational CR capacity exceeds the needs of its operating HDG lines. Consequently, the Plant would likely have at least some volumes of CR available for the merchant market even if its HDG lines were run at full capacity.

(1368) Additional CR capacity is also included particularly in the Galati Plant and in the Skopje and Liège Plants. With regard to the Divestment Businesses’ ability to exert a competitive constraint in Southern Europe, the Commission considers it necessary to ensure that in particular the capacity in Galati can serve customers on the merchant market in Southern Europe.

(1369) With regard to the ability of the Galati Plant to serve customers in Southern Europe, the Commission notes that some volumes of flat carbon steel are already supplied from the Galati Plant to Southern Europe. Overall, the Commission considers that the considerations discussed in recitals (1356) and (1357) with respect to HR apply to  CR as well. [Commercial information].

(1370) In conclusion, the Commission finds that the Commitments of 15 March 2018 are sufficient to fully remove the significant impediment to effective competition in relation to the production and supply of CR products to which the Transaction would give rise in the EEA. However, this is subject to an improvement of incentives for  the Galati Plant to serve customers in Southern Europe, for instance through the provision of distribution assets in that region.

13.3.4.4. Suitability to remove the significant impediment to effective competition in relation to horizontal non-coordinated effects in the production and supply of HDG products

(1371) In assessing the suitability of the Commitments of 15 March 2018 to remove the significant impediment to effective competition in relation to production and supply of HDG products, the Commission takes into account the following factors.

(1372) First, the Commitments of 15 March 2018 sufficiently remove the overlap in the operational HDG capacity brought about by the Transaction

(1373)  In Package 1, capacities are located at the integrated steelworks in Galati (Romania)  as well as in the satellite finishing plants in Piombino (Italy) and Skopje (FYROM). The majority of capacities in Package 1 are located in Piombino. In Package 2, capacities are located in Flemalle/Liège (Belgium) and Dudelange (Luxembourg).

(1374) As shown in Table 66, the combined currently operational capacity of the Commitments of 15 March 2018 reaches approximately […] Mt (920), which represents circa […]% of the overlap in HDG.

(1375)   [Commercial information].

(1376) The Commission observes that part of the operational capacity of the Divestment Businesses is located in Skopje (FYROM) and is thus outside the relevant  geographic market. Based on the information available to the Commission, it is not evident that this production site could effectively and competitively  supply  customers within the EEA and in particular in Southern Europe.

(1377) Nonetheless, the production capacity located in Skopje is notably small, only […] kt operational capacity. Hence, the outcome of the Commission’s assessment would not be altered even if that production site was considered unable to supply customers in the EEA as the production sites within the EEA have adequate operational capacities.

(1378) Second, the operational capacity and the actual production of the Divestment Businesses exceed Ilva’s current production and sales of HDG. In 2017, the Divestment   Businesses   produced   approximately   […]   Mt   of   HDG   of  which approximately […]  Mt in the EEA. At the same time,  Ilva produced […]  Mt in    the EEA.

(1379) Third, based on information provided by ArcelorMittal, the Divestment Businesses together produce all the grades of HDG currently produced by Ilva and their capabilities are also otherwise at least similar to those of Ilva’s even if the  capabilities of individual lines differ. This finding is also supported by the results of the market test in that the majority of both competitors and customers taking a position suggest that the capabilities of the Divestment Businesses are at least similar to those of Ilva’s. (921)

(1380) Fourth, the Commission considers that the Divestment Businesses are likely able to  act as significant competitive forces in the geographic areas where Ilva sells most of its HDG products and where the competitive interaction between the Parties is the strongest within the EEA.

(1381) The Commission recalls that the competitive interaction between the Parties is strongest in Southern Europe, in particular in Italy. In this respect the Commission observes that the only HDG facility included in the Divestment Businesses that is located in Southern Europe is the Piombino plant. Hence, while the operational capacity of the Piombino plant is noticeable – approximately […] Mt – only part of the overall HDG capacity is located in Southern Europe where Ilva sells most of its products. A small part of the remaining EEA HDG capacity of the Divestment Businesses is located in  Romania while the majority of it  is  located  in Belgium  and Luxembourg.

(1382) Nonetheless, the Commission recalls that, as indicated in Section 7.2, the geographic differentiation within the EEA is less pronounced with regard to HDG than, for instance, with regard to HR. Currently, Ilva sells a notable part of its HDG  production to customers beside Southern Europe, and in particular in […] where Ilva does not have production assets – as shown in Figure 35. Ilva makes over […]% (2016, approximately […] kt in volume) of its HDG sales to customers located in […]. The HDG facilities of the Divestment Businesses in Belgium and Luxembourg are at least equally suitably located to serve customers in […] and its surroundings as Ilva is.

(1383) In conclusion, the Commission finds that the Commitments of 15 March 2018 are sufficient to fully remove the significant impediment to effective competition in relation to production and supply of HDG products to which the Transaction would give rise in the EEA. This is, however, subject to improvements on technical matters such as questions on key personnel.

13.3.4.5. Suitability to remove the significant impediment to effective competition in relation to horizontal non-coordinated effects in the production and supply of EG products

(1384) In assessing the suitability of the Commitments of 15 March 2018 to remove the significant impediment to effective competition in relation to production and supply of EG products, the Commission observes that the considerations on HDG discussed in Section 13.3.4.4 largely apply to EG as well.

(1385) In particular, the Commission observes that the Commitments of 15 March 2018  would remove nearly all of the overlap brought by the Transaction: The   operational production capacity of the Dudelange EG plant is approximately […]% of that of Ilva’s. Based on information provided by ArcelorMittal, the Dudelange plant is also capable of producing all products Ilva is, a position that was supported in the market test by customers and competitors. (922)

(1386) All of the EG capacity included in the Divestment Businesses  is  located  in  Dudelange (Luxembourg) and thus none of it is in Southern Europe. However, the Commission considers that similar considerations apply in this respect as those discussed for HDG in Section 13.3.4.4. Moreover, while ArcelorMittal currently has notable EG production in Southern Europe (Spain), those production lines primarily produce products that Ilva does not and, consequently, ArcelorMittal produces products that are in competition with Ilva primarily outside Southern Europe prior to the Transaction.

(1387) In conclusion, the Commission finds that the Commitments of 15 March 2018 are sufficient to fully remove the significant impediment to effective competition in relation to production and supply of EG products to which the Transaction would give rise in the EEA. This is, however, subject to improvements on technical matters such as questions on key personnel.

13.3.4.6. Suitability to remove the significant impediment to effective competition in relation to horizontal non-coordinated effects in the production and supply of GS products (HDG+EG)

(1388)  The  Commission  has   also   assessed   the   suitability   of   the   Commitments   of 15 March 2018 under the potential overall GS  market  encompassing  both  HDG and EG.

(1389) As noted in recital (872), the overall GS market primarily consists of HDG with EG representing only a small part of the market. Moreover, the Commission’s conclusions explained in Section 13.3.4.4 with regard to HDG and those with regard to EG in Section 13.3.4.5 are similar.

(1390)  Therefore,  the outcome  of the Commission’s  assessment  of the Commitments  of  15 March 2017 would not be altered even if an overall GS market was considered instead of distinct markets for HDG and EG.

13.3.4.7. Suitability to remove the significant impediment to effective competition in relation to possible horizontal coordinated effects in the production and supply of flat carbon steel products

(1391) In Section 10, the Decision finds that the Transaction may lead to a significant impediment of effective competition due to horizontal coordinated effects. In particular, the Transaction would render price coordination in the EEA easier, more stable or more effective for firms that were already coordinating before the Transaction, at any rate, the Transaction would increase the likelihood that EEA producers of flat carbon steel are able to coordinate their behaviour and raise prices. The potential merger specific coordinated effects would in particular result from increased consolidation in the EEA and enhanced market transparency, enhanced potential for the merged entity to retaliate, and reduced scope for disrupting coordination, due to the removal of Ilva as an independent and price aggressive competitor and strengthened structural links between ArcelorMittal and Marcegaglia.

(1392) The Commission considers that the Commitments of 15 March 2018 are capable of removing the significant impediment to effective competition due to possible horizontal coordinated effects for the following reasons.

(1393) First, the Commitments can offset the disappearance of Ilva as an independent competitor by allowing new entry or significant expansion in the EEA for each of HR, CR and GS. This will reduce market consolidation and transparency, and allow the entry and/or strengthening the presence of EEA operators with the ability and the incentive to disrupt coordination.

(1394) Notably, the HR, CR and GS capacity in the Divestment Businesses is sizeable, and allows for significant new entry into the EEA and/or the significant strengthening of previously less important EEA operators. The purchaser(s) will have the incentive to compete aggressively to fill up both the existing and upgraded operational flat carbon steel capacity. In the first place, the Divestment Business(es) may lose a part of the customer base to ArcelorMittal or other established major EEA suppliers. According to the Notifying Party, the Divestment Business, including the integrated plants in Ostrava and Galati are cost competitive and therefore provide the Purchaser(s) with a viable platform to conquer market share by competing on price. In the second place, [commercial information]. In the third place, [commercial information]. Consequently, the Divestment Businesses provide significant incentives for the Purchaser(s) to compete aggressively.

(1395) Second, the Commitments reduce the merged entity's ability to retaliate against firms engaging in disruptive conduct. Due to the divestment of the Ostrava and Galati plants, ArcelorMittal's ability to exercise a credible retaliation threat in Central, Eastern and South Eastern Europe against aggressive competition in the rest of the EEA is significantly reduced.

(1396) Third, the Commitments remove envisaged reciprocal structural links between ArcelorMittal and Marcegaglia and thus further  strengthen  competition  for  CR  and GS. Absent Marcegaglia's stake in the acquired Ilva business of ArcelorMittal, Marcegaglia's financial performance will not depend on the performance of the merged entity. This will, in turn, reinvigorate Marcegaglia's incentives to compete aggressively with ArcelorMittal. Moreover, the removal of envisaged cross- shareholdings and related directorships will reduce the scope for exchange of sensitive commercial information between the two competitors. [Commercial information].

13.3.4.8. Suitability to remove the significant impediment to effective competition in relation to possible customer or input foreclosure concerns with regard to vertical links

(1397) The Commission finds that, in view of the final commitments entered into by the Notifying Party, it is not necessary to conclude as to whether the Transaction would lead to a significant impediment to effective competition in the internal market due to input foreclosure with regard to a number of vertical links.

(1398)  These are the vertical links between the production and supply of different finished  flat carbon steel products (Section 9.13); the vertical links between the production and supply of finished flat carbon steel products, and the distribution of such  products (Section 9.14); the vertical links between the production and supply of finished flat carbon steel products, and of carbon steel tubes (Section 9.15); the vertical links between the production and supply of finished flat carbon steel products, and the production and supply of various other downstream products (Section 9.16).

(1399) As explained in Sections 13.3.4.2 to 13.3.4.7 the Commission considers that the Commitments of 15 March 2018 are capable of removing the significant impediment to effective competition of the Transaction due to possible horizontal non- coordinated and coordinated effects.

(1400) The possible concerns related to input foreclosure with regard to the vertical links are all related to the horizontal overlap between the Parties. As the Commitments of 15 March 2018 removes the significant impediment to effective competition due to the horizontal overlap between the Parties the Commission then also the possible concerns related to the vertical links are removed.

13.3.5.        Viability, competitiveness and suitable purchaser

(1401) The Commitments of 15 March 2018 did not sufficiently ensure the competitiveness and viability of  the  Divestment  Business  as  proposed  in  the  Commitments  of  15 March 2018. This is for the following reasons.

(1402) First, the Commitments of 15 March 2018 did not include sufficient arrangements for the supply of HR to CR and GS plants to complement the purchasers' own HR supplies if necessary to feed the downstream finishing plants of the Divestment Businesses.

(1403)   The Commission recalls that HR is used as the main input in the production of CR.   In addition, all GS production eventually depend on HR supplies either directly (for certain HDG products) or through the intermediate production of CR substrate for them (HDG and EG).

(1404)   [Commercial information].

(1405)   [Commercial information].

(1406)   [Commercial information].

(1407)   [Commercial information]. (923)

(1408) The Commission further recalls that, as explained in this Decision (see for instance Section 9.7.2.2), re-rollers may not be as competitive as integrated producers such as Ilva or ArcelorMittal. Hence, it cannot be excluded that if the supply of HR to Package 2 is significantly dependent on merchant market supplies, the Divestment Businesses may not be able to exert similar competitive constraints on ArcelorMittal as Ilva can prior to the transaction.

(1409)   [Commercial information].

(1410) Therefore, the Commission considers that the lack of sufficient arrangements for vertical integration and HR supply would impair the viability and competitiveness of the Divestment Business as a standalone business.

(1411) Second, the Commitments of 15 March 2018 did not include sufficient arrangements for the supply of HR to CR and GS plants to allow for smooth business transition following the divestment, and until a suitable purchaser could reproduce adequate supply flows to the finishing lines in the Divestment Business.

(1412) The Commission notes that the Commitments of 15 March 2018 did not include any transitional arrangements providing HR to the divestment businesses.

(1413)   […].

(1414) […]. A number of customers responding to the market test have indicated that a  change in the substrate could affect their willingness to purchase from those facilities (924) even if the majority of customers and competitors alike accepted that the purchaser of the assets could in general contribute its own HR for the production of CR and GS within the Divestment Business. (925)

(1415) Therefore, the Commission considers that the lack of any transitional supply arrangement would impair the viability and competitiveness of the Divestment Businesses.

(1416) Third, the Commitments did not offer sufficient certainty that the Ostrava plant could be viable and competitive on a standalone basis, in particular due to the proposed carve out of the on-site companies as well as related downstream businesses which currently contribute to the operation of the Ostrava plant.

(1417) Fourth, the Commitments did not sufficiently specify the purchaser criteria or  stipulate the open and non-discriminatory nature of the sales process.

(1418) In the first place, preference should be given to a single purchaser of the Divestment Businesses. The objective of the remedy should be to reproduce the constraint exerted by Ilva as a large and integrated player with both primary steelmaking as  well as sizeable finishing lines including modernised facilities in Novi Ligure and Genova. The remedy should replicate the significant scale effects of Ilva's  pre- merger configuration to the fullest. [Commercial information].

(1419)   [Commercial information].

(1420) Additional emphasis on the requirement that the proposed Purchaser ought to have access to sufficient HR to viably operate the business and having material volumes of owned HR capacity in the EEA to feed the CR and GS lines at full capacity reduces the risk that the Commission would be presented a purchaser that fails to meet such basic requirements. For the Notifying Party, the extended purchaser requirements would provide additional guidance which purchasers are likely to qualify as acceptable buyers of the divestment businesses.

(1421) In the third place, considering the importance of the purchaser having established production and supply of HR flat carbon steel, and in light of the findings of possible coordinated effects, the Commission considers it important for the Notifying Party to ensure, and for the monitoring trustee to check, that all potential suitable buyers are given the opportunity to be involved on equal footing in the buyer selection process. In doing so, the Commission aims to minimise the risk that interested and viable market operators are excluded from acquiring the divestment business and strengthening their market positions vis-à-vis the Parties to the Transaction or enter the European markets de novo.

(1422) Sixth, the Commission considered that the Commitments were not sufficiently effective on a number of other issues, such as the [commercial information] and the scope and identity of key personnel.

13.3.6. Conclusion

(1423) Based on its own assessment of the Transaction and the responses received in the course of the market test, the Commission concludes that, the Commitments of 15 March 2018 could generally be suitable to address the competition concerns identified by the Commission. However, in view of  the  reasons  set  out  in  Sections 13.3.3 to 13.3.5 above, the Commitments of 15 March 2018 did not constitute a viable and competitive business capable of replicating the constraint lost because of the Transaction.

13.4. Final Commitments

13.4.1. Description

(1424) On 11 April 2018, the Notifying Party submitted revised commitments (the 'Final Commitments'). The full text of the Final Commitments is attached as Annex 2 to  this Decision.

(1425) The Final Commitments consist of the divestiture of all the assets of the Divestment Business as described in  the  Commitments  of  15 March  2018  as  listed  in  Section 13.3.1, and also including (the 'Divestment Businesses'):

(1) Distribution assets in France and Italy that will be sold within Package 1, and  as such will be attached to the Piombino/Galati Business;

(2) Ostrava's tubular business (Package 3).

(1426) As already foreseen in the Commitments of 15 March 2018, [commercial  information].

(1427)   Further changes in the Final Commitments as compared to the Commitments of 15 March 2018 include:

(1)   The requirement to hold an open sales process in which all potentially suitable purchasers are able to participate on an equal and non-discriminatory footing and to keep the Commission and the monitoring Trustee regularly informed about its progress;

(2)   A transitional supply agreement for HR for the Piombino plant and the Liège/Dudelange Business to feed their GS lines, [commercial information];

(3)   [Commercial information].

(4)   The commitment to transfer the service agreement with the energy company that provides utility services to the Ostrava plant to the Purchaser of Package 3; and the option for the Purchaser of package 3 to obtain either ArcelorMittal's stake in the energy company that provides utility services to the Ostrava plant or the underlying assets.

(5)   The commitments provide clarifications that the pickling lines in the Liège plant (Package 2) are included;

(6)   [Commercial information].

(7)   As to the key personnel, additional persons were included for Piombino, Galati, Skopje, Liege, Dudelange and Ostrava, covering the roles of overall plant management, Automation & Industrial IT Support, plant safety, quality and maintenance, energy, legal, finance and accountancy.

13.4.2.        The Commission's assessment

(1428) Having assessed the Final Commitments, the  Commission  has  reached  the conclusion that, due to the inclusion of the changes described in Section 13.4.1, the Transaction would no longer give rise to a significant impediment of effective competition in the EEA.

(1429) First, by requiring that a potential purchaser of Package 2 on a stand-alone basis has sufficient access to HR within the EEA, the remedy ensures that the downstream assets in that Package can be operated competitively and viably due to benefits from vertical integration, notably cost structure and security of supply.

(1430) [Commercial information]. ArcelorMittal and Ilva are both integrated players active throughout the flat carbon steel supply chain and operate a business model that is different from that of a re-roller not active in basic steel-making. […].

(1431)   [Commercial information].

(1432) Third, the inclusion of the distribution assets in Southern Europe to Package 1 further enhances the incentives to supply that area from the Galati plant. This is in line with certain market test observations. Specifically, several customers indicated that such distribution outlets are required to increase incentives for the Purchaser of Package 1 to supply in Southern Europe where the Parties activities overlap the most. (926)

(1433) Fourth, the inclusion of the transitional supply agreements to provide HR to the Piombino Plant and the Liège/Dudelange Business addresses the concern of customers who indicated this to be needed for a smooth transition to new supply sources. This would provide for the short term viability and competitiveness of the Divestment Businesses by ensuring non-interrupted operations and customer relationships following the divestment or would complement the purchasers' own HR supplies to feed the Divestment Businesses' downstream assets. (927)

(1434) Fifth, the Final Commitments endorse an open and non-discriminatory sales process. Considering the Commission's findings in the market investigation on non- coordinated and coordinated effects in the production and supply of flat carbon steel, it is necessary that the purchaser is fully independent of the Notifying Party. Therefore, the independence criteria will be enforced strictly, and to that extent, the monitoring Trustee will be checking whether all potential suitable buyers are given the opportunity to participate in the purchase process on an equal footing and non- discriminatory basis. The Commission considers this likely to guarantee that viable market operators are not excluded from the opportunity of entering the European markets or to strengthen their market position vis-à-vis the Parties to the Transaction.

(1435) Sixth, the Commitments improve the competitiveness and viability of the Ostrava Business.

(1436) In the first place, the inclusion of ArcelorMittal's tubular business in  Ostrava  enhances the viability of Package 3, [commercial information].

(1437) In the second place, the Final Commitments now contain the option for the purchaser to obtain ArcelorMittal's stake in the energy company that provides utility services to this plant or to obtain ArcelorMittal's stake in the underlying assets. This will allow the Purchaser to have increased control over this input and possibly mitigate energy costs and benefit the viability and competitiveness of the Ostrava business.

(1438)   Seventh, [commercial information].

(1439) Eighth, the expansion of the scope of key personnel addresses the Commissions reservation on the Commitments of 15 March 2018 as to the required personnel necessary to ensure the viability and competitiveness of the Divestment Businesses.

13.5. Conclusion

(1440) 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.

14.  CONDITIONS AND OBLIGATIONS

(1441) 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.

(1442) The fulfilment of a measure that gives rise to a structural change of the market is a condition, whereas the implementing steps which are necessary to achieve that 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 will not or no longer be 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.

(1443) This Decision should be made conditional on the full compliance by the Notifying Party with Section B (including Schedules 1 to 3 and their sub-schedules) and C of the Final Commitments. All other sections of the Final Commitments should be made obligations within the meaning of Article 8(2) of the Merger Regulation. The full  text of the Final Commitments is attached as Annex 2 to this Decision and forms an integral part thereof.

HAS ADOPTED THIS DECISION:

Article 1

The notified operation whereby ArcelorMittal acquires sole control of the Ilva assets within the meaning of Article 3(1)(b) of Regulation (EC) No 139/2004 is compatible with the internal market and the EEA Agreement.

Article 2

Article 1 is subject to compliance with the conditions set out in Sections B and C of Annex 2 to this Decision.

Article 3

ArcelorMittal 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: ArcelorMittal S.A.

24–26, Boulevard d‘Avranches

1160 Luxembourg

Luxembourg

Done at Brussels, 7.5.2018

CASE M.8444 – ARCELORMITTAL/ILVA

ANNEX 1 TO THE COMMISSION DECISION

QUANTITATIVE ANALYSIS

1.  INTRODUCTION

(1)                                      The Commission carried out a quantitative analysis to investigate the effects certain events on sales by different market participants. Specifically, the analysis estimates the losses or gains of HR sales shares by market participants in Southern Europe, and most notably by ArcelorMittal and imports, resulting from the imposition of the capacity cap on Ilva in late 2012. The analysis also examines the shifts in sales  shares during the period 2015-2016 when imports from certain non-EEA countries, which were subsequently found to be dumping, peaked in the EEA, and when Ilva was subject to further capacity caps.

(2)                                      The analysis further estimates the elasticity of the HR sales shares from domestic EEA suppliers to imports with respect to the Southern European price, controlling for factors that determine the availability of imports. That is to say, the percentage points shifts in HR sales shares between EEA suppliers and imports following one percent increase in Southern European HR price.

(3)                                      The results of the Commission's quantitative analysis corroborate the qualitative evidence gathered during the Commission's in-depth investigation regarding the closeness of competition between ArcelorMittal and Ilva. Following the imposition  of the capacity cap on Ilva, the analysis finds that ArcelorMittal gained a sizeable share of the HR market in Southern Europe and in Italy, implying a high diversion ratio from Ilva to ArcelorMittal. This is consistent with ArcelorMittal internal documents described in Section 9.4.2.6 of the Decision showing […], and indicates a strong competitive interaction between the two companies.

(4)                                     The analysis also finds that the share of imports has increased substantially in 2015, while the share of Ilva has decreased with no strong shift in Arcelor Mittal's share. These shifts are likely due to the combined effect of (i) dumping of volumes in the EEA by producers of certain non-EEA countries, and (ii) further capacity reductions at Ilva. However, as these events happened simultaneously, it is not possible to disentangle the two effects, and hence it is difficult to interpret those results from the perspective of closeness of competition.

(5)                                   The analysis further sheds light on the competitive constraint exerted by imports on domestic EEA suppliers and their ability and incentives to constrain the pricing behaviour of the merged entity post-Transaction.

(6)                                   First, the incentives of non-EEA suppliers to increase volumes towards Southern Europe crucially depend on the domestic prices in the exporting countries. This makes of imports a volatile source of competitive constraint in Southern Europe, as the constraint from imports will depend on factors beyond those affecting  competition in Southern Europe.

(7)                                  Second, the constraint exerted by imports, in terms of HR volumes that could be directed towards Southern Europe, depends, all else equal, also on Southern European HR prices. However, the extent to which customers would reduce their HR purchases from domestic EEA suppliers and shift to imports following an increase in Southern European market prices does not appear to be such as to eliminate the merged entity's incentives to increase prices post-Transaction.

(8)                                 As clarified in the Letter of Facts, (1) the Commission's quantitative analysis aims to estimate to what extent EEA customers would be willing to reduce their purchases from EEA suppliers and, consequently, increase their purchases from imports.  Hence, the aim of the Commission's quantitative analysis is to gather insights on the elasticity of the demand faced by EEA producers taking into account customer switching to imports. (2)

(9)                                 Contrary to the Notifying Party's arguments, this approach is not "indirect". (3) The extent to which the elimination of competition between the Parties will provide the merged entity with an incentive to raise price post-Transaction will depend, all else equal, on the competitive constraints from competitors in the EEA as well as on the competitive constraint imposed by imports. The estimation of the residual demand elasticity faced by domestic EEA suppliers hence provides a direct measure of the competitive constraint from imports on EEA suppliers. The constraint on the  merged entity from its rivals in the EEA is not quantified by this analysis. (4)

(10)                             The quantitative analysis relies on data from the period December 2005-June 2017. As anti-dumping measures were only introduced in October 2016, it is not possible, in the context of this analysis to reliably estimate how these trade measures have affected the degree by which imports react to price changes in Southern Europe. However, other evidence indicates that trade measures have reduced the  effectiveness of imports in constraining prices in the EEA, compared to the period covered by this analysis, as indicated in Section 9.4.5.7 of the Decision.

(11)                           In its Reply to the SO and its Reply to the Letter of Facts, the Notifying Party made several criticisms in relation to the Commission's quantitative analysis, which are addressed in the relevant Sections of this Annex. Overall, the Commission considers that such criticisms do not fundamentally change its conclusions.

(12)                          This Annex describes the methodology of the Commission's quantitative analysis, and discusses its results. The Annex is organized as follows. Section 2 describes the events affecting the HR flat steel market in the EEA, and in particular Southern Europe, that are relevant for the analysis. Section 3 discusses the incentives for non-EEA suppliers to export HR volumes into the EEA, and how this is taken into account in the analysis. Section 4 and Section 5 describe, respectively, the data used in the analysis and its methodology. Results and robustness tests are presented and discussed in Section 6 and Section 7. Finally, Section 8 concludes.

2. THE EVENTS AFFECTING THE SOUTHERN EUROPEAN FLAT STEEL INDUSTRY

(13)                         The period under investigation was characterised by several events affecting competition in the HR market in Southern Europe. Specifically, the imposition of a capacity cap on Ilva's production plant in Taranto, the dumping volumes of HR from non-EEA countries, and the imposition of anti-dumping duties against those countries. This Section provides a brief description of those events with a particular emphasis on the aspects that are relevant for the quantitative analysis.

2.1. The Imposition of the Ilva Capacity Cap

(14)                          For the purpose of the analysis, the imposition of a capacity cap on Ilva amounts to two subsequent events that caused a reduction in Ilva's capacity and production at  two different points in time: in late 2012 (the '2012 Ilva Cap'), and in early 2015    (the '2015 Ilva Cap').

(15)                         Due to environmental issues affecting Ilva's plant in Taranto, throughout 2012  several court judgements and law decrees required Ilva to mothball its blast furnace 1 (BF1) in November/December 2012. This reduced the molten iron capacity of Ilva  by approximately […].

(16)                          In March 2015, also Ilva's biggest blast furnace (BF5), with a capacity of […], was mothballed, while a few months later, in August 2015, BF1 was restarted. As BF5 had a higher capacity than BF1, the net effect by the end of 2015 was a further reduction of Ilva's molten iron capacity by […]. Due to the 5 months delay separating the mothballing of BF5 and the restarting of BF1, the negative effect on Ilva's capacity and production was even greater during 2015.

(17)                        The effects of the 2012 and 2015 Ilva Cap are visually summarized by Ilva's HR production over time depicted in Figure 1.

Figure 1: Ilva HR production from 2011 to 2016

[…]

Source: Commission's computation based on Notifying Party, Response to RFI 19, Annex 18.1

(18)                            From 2012 to 2013, the mothballing of BF1 resulted in a steep decrease in Ilva's HR production. Production remained stable in 2014, to fall sharply again in 2015 due to the mothballing of BF5. The HR production then increased in 2016 following the restart of BF1, although it reached a lower production level than in 2014 due to the difference in capacity between BF5 and BF1. […].

(19)                           The two subsequent capacity reductions are therefore binary events at two different points in time. A more gradual reduction of Ilva's sales is likely explained  by material in stock at the time of the mothballing of BF1 that was delivered over time thereafter.

(20)                           From an econometric perspective, the two capacity reductions at Ilva's Taranto plant can be considered exogeneous events, as they were not planned by Ilva itself and were not foreseen by the market (as, for example, a temporary plant closure to carry out maintenance might be). (5)  The 2012 capacity reduction was the result of orders  by the Italian courts and law decrees by the government; whereas the 2015 capacity reduction, although it may have been planned in advance, was highly uncertain, as shown  by  ArcelorMittal's  internal  documents  discussed  in  Section 9.4.2.6  of   the Decision.

2.2. Dumping period and Anti-Dumping measures

(21)                           Following the complaints of Eurofer, the European steel producers association, during 2016 the Commission (DG Trade) opened two anti-dumping (AD) investigations against several countries to assess whether the imported HR volumes should be considered as dumping.

(22)                           In October 2016, the Commission imposed preliminary AD measures against China. The Commission found that within  the  investigation  period  from  January  2015, the HR volumes from China were dumping volumes. Final AD measures were imposed against China in April 2017. (6)

(23)                          At the same time, in July 2016, also an investigation against Russia, Iran, Ukraine, Brazil and Serbia started. In January 2017, a registration requirement was imposed against HR volumes from Russia and Brazil. Finally, the Commission found that during  the  investigation  period  from  July  2015,  HR  imports  from  Russia,  Iran, Ukraine and Brazil (but not Serbia) constituted dumping and, in October 2017, final AD measures were imposed against those countries. (7)

(24)                         The effects of the AD measures on imported HR volumes to Europe is visible in Figure 2.

(25)                            Following the imposition of preliminary AD measures in October 2016 (the blue dashed line in the figure), imported HR volumes from China dropped to virtually zero. Similarly, the imposition of a registration requirement to Russia and Brazil       in January 2016 (the red dashed line in the figure), (8) significantly decreased the imported volumes from Brazil, Russia, Iran and Ukraine.

(26)                           The effects that dumping volumes and AD measures had on competition in Southern Europe may have had different intensities at different points in time, depending, inter alia, on the exporting countries involved. However, for the purpose of the quantitative analysis it is appropriate to follow the timing of the investigation against China, as it encompasses also the investigation against the other exporting countries and its effects. Therefore, two periods are relevant: the period as of which dumping volumes were hitting the EEA from January 2015, and the period following the first AD measures, from October 2016.

(27)                           The fact that the dumping period and the 2015 Ilva Cap happened simultaneously (from early 2015) implies that the effects of the two events cannot be identified separately in the quantitative analysis. Only the combined effect of the two events can be estimated. This issue and its implications will be discussed in more detail in Section 5 of this Annex.

3.  THE IMPORTANCE OF RELATIVE PRICES FOR IMPORTS

(28)                          The relative price between domestic HR market prices in Southern Europe and in China is an important driver of HR imports to Southern Europe. (9)

(29)                         As HR prices in Southern Europe increase relative to domestic Chinese prices, exporters have higher incentives to export HR volumes to Southern Europe. Conversely, as domestic Chinese HR market prices increase relative to prices in Southern Europe, exporters have lower incentives to export HR volumes to Southern Europe.

(30)                        Figure 3 illustrates the intuition and suggests that relative HR prices explain well the movement in the imported HR volumes in Europe.

(31)                       The blue line in the figure represents the relative domestic HR price in Southern Europe and China, lagged by three months. (10) An increase in the relative price means that Southern Europe becomes more attractive relative to the domestic market in China. It is necessary to consider the lagged value of the relative price because the underlying price series are recorded when the order for a given HR quantity is made, whereas the volumes are recorded when the delivery is made. (11) The red line in the figure represents the total imported HR volumes in Europe. To improve readability, both series are displayed as moving averages.

(32)                           Figure 3 shows that the movement in the relative HR price appears to explain to a large extent the movements in the imported HR volumes to Europe. In particular, the high volume of imports in 2008, followed by the drop in 2009 is mirrored by a similar movement of the relative HR price. The same occurred with the increase in imports in 2010 and 2011, followed by the drop in 2012/13.

(33)                          The fit between relative HR prices and imported HR volumes is relatively strong  until 2015, then the steep increase in imported volumes is not accompanied by an increase in relative prices of the same magnitude. This is likely due to the period characterised by dumping volumes coming to the EEA.

(34)                         The following conclusions can be drawn from Figure 3.

(35)                         First, at least until the dumping period (i.e. until 2015) imported HR volumes appear to react to changes in relative prices. On the one hand, this suggests that imports would increase following a price increase in Southern Europe, all else being equal. On the other hand, it also suggests that imported HR volumes to Europe depend on factors beyond those affecting the steel industry in Europe, most notably the factors affecting domestic HR market in the exporting country. This would qualify imports' behaviour as opportunistic and, therefore, the competitive constraint that they represent in the HR market in Europe as volatile.

(36)                       Second, for the purpose of the quantitative analysis it is important to take the relative prices into account when analysing an event such as the introduction of the Ilva Cap. This is because movements of imported volumes over time may be erroneously associated to the event under study, while, in fact, they are due to factors beyond the HR market in Europe that affect the relative price.

(37)                       Third, for the purpose of analysing the extent to which EEA customers are willing to switch their purchases from EEA suppliers to imports (the elasticity of the demand for HR supplied by EEA producers, or the share of these producers demand), it is important to take account of factors that affect the availability of imports in the EEA, such as the domestic Chinese prices, the dumping period and the subsequent AD measures. This ensures that the analysis is done on an "all else being equal" basis, i.e. holding constant the availability of imports to EEA customers.

4.   DATA

(38)                          The analysis uses data from several sources to construct a dataset at monthly level  for the period December 2005 – June 2017 for all countries within EU28.

(39)                         Table 1 provides an overview of the data used and their sources.

(40)                       The HR volume figures include sales to re-rollers (for EEA suppliers and for imports from non-EEA producers), and are used to compute monthly sales shares that will be used in the analysis.

(41)                      The price series (in EUR/t) are market prices recorded by SBB through interviews with customers and traders in which quotes are collected and then averaged to form a price. (14)

(42)                      As mentioned in paragraph (31), the price quotes are recorded at the time of the order, whereas volumes are recorded at the time of the delivery. Therefore, in the analysis the Commission uses the lagged value of three months of the price series to correct for the difference in timing between order and delivery.

(43)                     Raw material prices include the input prices used in the primary steel production, most notably coal, iron ore and scrap metal.

(44)                     Seasonally adjusted GDP and worldwide crude steel consumption are used (in first difference (15)) to control for the economic cycle and macro-economic trends (such as the economic crisis of 2008/2009).

(45)                    Finally, further robustness tests presented in the Letter of Facts include the Industrial Price Index (IPI) to further control for demand factors over time.

5.  METHODOLOGY

(46)                    The methodology adopted by the Commission is a regression analysis of sales shares on prices, dummy variables determining the events, and control variables. The sales shares are those of ArcelorMittal ("AM"), Ilva, imports, and the aggregate of Other EEA suppliers.

(47)                  Specifically, the regression has the following form:

ssMareict = a + QD + yPt + yGDPct + qCt (+$Ic + ðIPIct) + cict         ( 1 )

(48)                  Where:

(a)   sshareict is the HR sales share of firm i (AM, Ilva, Imports, or other EEA suppliers) in country c (depending on the specification there can be several countries or just one) at time t (month);

(b)   the matrix D includes the dummy of the events. (16) Specifically, (i) a dummy equal to one from November 2012 onwards capturing the 2012 Ilva Cap, (ii) a dummy equal to one from January 2015 onwards, capturing the 2015 Ilva Cap as well as the period when imports were considered dumping, and (iii)  a dummy equal to one from October 2016 onwards capturing the period post AD measures;

(c)   the matrix Pt includes the natural logarithm of domestic HR prices in Southern Europe and China taken at the third lag for  the  reason  explained  in  paragraph (31); the subscript t underlines that the domestic market prices do  not vary across firms or countries;

(d)   GDPct is the first-difference of the GDP of country c at time t;

(e)   Ct is the first-difference of the worldwide consumption at time t;

(f)   Ic includes country-specific fixed effects in the specifications where there are more than one country;

(g)   IPIct is the (log) industrial price index of country c at time t, which is used in further robustness tests discussed in Section 7.4;

(h)   cict is the error term.

(49)                               The regressions are presented in groups of four, one for each company. This is because the analysis exploits the mathematical property whereby, by regressing the sales shares of the four firms on the same variables, the sum of the same coefficient across the four regressions is constrained to be equal to zero. (17) This implies that the interpretation of the regressions' coefficients will be the effect of a given variable on the sales shares of one firm, taking into account the effect of the same variable to the other firms.

(50)                               The coefficients þ measure the change in percentage points of the sales shares of one firm following the events abovementioned, taking into account the change in the  sales shares of the other firms. The þ associated to the 2012 Ilva Cap (þ1) is of particular interest for the Commission's quantitative analysis because it gives a direct measure of the substitution between Ilva's and competitors' HR products. From the outset, þ1 is expected to have a negative sign in the Ilva regression, and a positive sign in the regressions of the other firms. That is to say, after the imposition of the 2012 Ilva Cap, Ilva is expected to have lost sales shares to the benefit of competitors. The magnitude of the þ1 coefficients for the competitors will then be a measure of closeness of competition.

(51)                               It is not possible to  associate  the  same  inference  to  the  coefficients  related  to  the 2015 Ilva Cap (þ2). This is because in 2015-2016 two simultaneous events took place: a further reduction in the Ilva capacity (i.e. the 2015 Ilva Cap) and the period of dumping HR volumes from outside of the EEA. The regression analysis is not able to disentangle the effects that each of those two events had on the sales shares of the market participants. The þ2 coefficients only capture the joint effect of the two events. From the outset, þ2 is expected to have a positive sign in the Imports regression due to the increased volumes resulting from the dumping practices, and a negative sign in the Ilva regression due to the further decrease in Ilva capacity. However, the sign of þ2 in the ArcelorMittal regression or in the regression of the Other EEA Suppliers is not clear. On the one hand, they may gain sales shares due to the reduction of Ilva's sales shares. On the other hand, they may lose sales shares to Imports. Hence, the interpretation of the þ2 coefficients in terms of closeness of competition may be confounded by the opposite effects that the two simultaneous events have on the sales shares of ArcelorMittal and of the Other EEA suppliers. (18)

(52)                             Finally, the þ coefficients associated to the period post AD measures (þ3) quantifies the change in sales shares following the imposition of (preliminary) AD measures. From the outset, þ3  is expected to have a negative sign on the Imports regression,   and a positive sign on the other market participants' regressions. As it will be shown in Section 6 discussing the results, in some specifications this is not the case and the þ3 coefficient has a positive sign in the Imports regression, implying that on average imports increased post-AD measures. This suggests that there was a substitution effect of imports from countries not affected by AD measures replacing the imports from countries affected by AD measures. However, the dataset used in the quantitative analysis has only limited information post-AD measures, which may undermine the reliable estimation of the þ3  coefficients.

(53)                            In its Reply to the SO, (19) the Notifying Party argued that the dummies identifying the dumping period and the post-AD measures period should not be included because the effects of those events on the sales shares in the EEA should be already captured by the domestic price variables. (20)  The Commission considers that, as already shown in Figure 3, the steep increase in imported HR volumes to the EEA from the beginning of 2015 appears not to be explained by an increase in the relative domestic prices in Southern Europe and China. Therefore, controlling for the dumping period and the subsequent AD measures is important to properly identify the volume diversion from Ilva to the competitors following, at least, the first imposition of the capacity cap to Ilva in 2012. (21)

(54)                          The coefficients associated to the HR domestic Southern European market price and the Chinese market price (the y coefficients) are also of interest for the quantitative analysis, in particular those relating to the Imports regression. As the prices are measured in natural logarithms, the y coefficients capture the percentage point change in sales shares caused by a 1% increase in domestic prices, all else equal. Specifically, the coefficient associated to the HR domestic Southern European  market price (y1), measures the percentage point change in sales shares for a 1% increase  in  Southern  European  prices,  all  else  equal.  Likewise,  the    coefficient associated to the HR domestic Chinese market price (y2) measures the percentage point change in sales shares for a 1% increase in Chinese prices, all else equal. The y coefficients in the Imports regression are particularly relevant for the analysis because  they  measure  the  elasticity sales share  relative  to the  Southern European prices and to the Chinese prices. Such measures may provide insights on the constraining effect of the Imports on the merged entity's post-Transaction.

(55)                        Lastly, the coefficients associated to the first-differences of GDP, worldwide consumption, and (log) IPI (5, q and ð) measure the change in sales shares given by the growth of GDP and worldwide consumption, and the percentage change in IPI. Those variables are used to control for shifts in demand, that is to control for the effects that macro-economic trends or macro-economic events may have on the demand of HR. Similarly, the country fixed effects Ic control for country-specific unobservable factors that do not vary in time. The analysis will not focus on those measures.

(56)                        In the Reply to the SO, the Notifying Party argued that the fitted value of the Commission's quantitative analysis does not precisely reproduce the evolution over time of the sales shares of Ilva and imports, and this leads to an underestimation of the sales diversion effect between Ilva and imports. In this respect, the Commission notes the following. First, the Notifying Party itself does not attempt any correction for this criticism in its subsequent analyses in the Reply to the SO. Second, it is common in a regression framework to have the fitted values not perfectly resembling the underlying series. The Notifying Party's criticism essentially points to the R2 in the Commission's regressions being lower than 100%, which is the norm in regression analysis. The Commission considers that the level of the R2 in the regressions of its quantitative analysis is generally at an acceptable level. Third, the discrepancy between fitted values and actual series appears to be mostly present in the period 2006-2008, which has been excluded in one of the Commission's robustness tests without any significant change in the results. Third, the identification of the events of interest in the Commission's quantitative analysis averages the fitted values over a relatively long period of time prior to the imposition of the 2012 Ilva Cap. As a consequence, to the extent that the fitted values underestimate and overestimate the actual sales shares over time, the average discrepancy is diluted. Overall, the Commission does not consider the Notifying Party's criticism to significantly affect the quantitative analysis.

(57)                        The canonical assumption that the error term is i.i.d. (identically and independently distributed) is likely to be violated in Equation (1) due to the fact that HR sales  shares are likely auto-correlated over time. For this reason, the Commission uses heteroscedasticity and auto-correlation robust (HAC) standard errors.

5.1. Endogeneity and Instrumental Variables Regressions

(58)                          A potential issue that may arise in Equation (1) is endogeneity. Endogeneity problems arise when explanatory variables (i.e. variables in the right-hand side in Equation (1)) are correlated with the error term c. Endogeneity issues may be due to  a  variety  of  reasons,  including  (i)  omitted  variables,  (ii)  measurement error, (iii) simultaneous causality. Estimates based on ordinary least squares (OLS) are biased in presence of endogeneity.

(59)                          In Equation (1), the HR price in Southern Europe and the sales shares may be simultaneously determined, hence giving raise to the simultaneous causality problem. One method to investigate and to solve the potential endogeneity issue is to use instrumental variable ('IV') estimation. (22),(23)

(60)                          In essence, the IV estimation makes use of the variation caused by other variables (called the 'instrumental variables' or the 'instruments') on the endogenous explanatory variable to investigate the causal relation between the endogenous variable (i.e. the HR price in Southern Europe) and the independent variable (i.e. the sales shares).

(61)                         To be used within the framework of instrumental variable estimation, an instrument has to be valid, that is to say, it has to respect the following conditions:

(a)   Relevance: The instrument has to be correlated with the endogenous explanatory variable.

(b)   Exclusion: The instrument has to be correlated with the independent variable only through the endogenous explanatory variable, and should not be  correlated with the error term c (i.e. they should not be endogenous themselves).

(62)                        Moreover, if the correlation between the instrument and the endogenous variable is small, the instrument is called weak. This may introduce a bias in the estimation, the size of which is generally inversely related to the correlation between the instrument and the endogenous variable.

(63)                        The Commission applied the IV estimation by using as instruments the raw material prices, specifically the iron ore price and the scrap metal price, which are strongly correlated with the domestic HR price in Southern Europe (i.e. they are not weak instruments). Raw material prices are broadly considered by the economics literature as valid instruments for the estimation of structural components of demand, and are widely used in peer reviewed publications. Therefore, in the Commission's view, those instruments are valid instruments.

(64)                       The Commission then tested the endogeneity of the domestic HR price in Southern Europe in each regression using a statistical test for endogeneity. (24) The test provided no indication that the Southern European HR market price is endogenous relative   to each of the HR sales shares of the market participants. (25) This result can be explained by the following reasons: (i) the domestic Southern European HR market price is, indeed, exogenous relative to each market participants' sales shares, (ii) the regressions control for the source of endogeneity, hence eliminating it, and (iii) there is an endogeneity bias, but its magnitude is small. In particular, in relation to the  latter reason, the Commission considers that it is plausible that the endogeneity bias, if any, is small. This is because the HR sales shares of each individual market participant are unlikely to fully determine the market price.

(65)                      The result of the statistical test generally suggests the use of OLS regressions rather than IV regressions, because OLS regression are more "efficient". However, in the following Sections the Commission presents both results based on OLS regressions and IV regressions.

6.  RESULTS

(66)                      This Section presents the results of the Commission's quantitative analysis, and its robustness checks. As competition between ArcelorMittal and Ilva is particularly intense in Southern Europe, the Commission focuses its quantitative analysis on Southern Europe (26) and on Italy.

6.1. OLS Regressions Results

(67)                      Table 2 presents the results of the OLS regression in Southern Europe. The four numbered columns are each associated with a regression of the sales' shares of one market participant. Specifically, the first column displays the Ilva regression, the second column displays the ArcelorMittal regression, the third column displays the Imports regression, and the fourth column displays the Other EEA Suppliers regression. It should be recalled that the coefficients for the same variable in the four columns sum to zero. In particular, the sum of the coefficients in regressions of Ilva, ArcelorMittal and Other EEA Suppliers, which represent the domestic EEA suppliers, will be equal to the coefficient in the Imports regression, with the opposite sign. For simplicity, in describing the coefficients on the price variables, the Commission will refer to the Imports regression, as the percentage points change in sales shares of domestic EEA suppliers that is diverted to imports.

(68)                     The variables on the left-hand side of the table are those described in Section 5. In order, L3.lpdom_SEU_HR is the natural logarithm of the domestic HR price in Southern Europe taken as third lag. Similarly, L3.lpdom_IMPCH_HR is the natural logarithm of the domestic HR price in China taken as third lag. Then there are the three dummy variables representing the events, ilvacap2012 is the dummy variable representing the imposition of the Ilva Cap from November 2012 onwards, ilvacap2015_dumping is the dummy variable representing the 2015 Ilva Cap and the dumping period, and AD is the dummy variable representing the period post-AD measures. Finally, D.gdp and D.ww_consumption are the first-difference of GDP and worldwide crude steel consumption.

(69)                        Examining the coefficients of the dummy variables, the imposition of the Ilva Cap   in 2012 (ilvacap2012) had a negative effect on Ilva's HR sales share in Southern Europe of […].p. In  the  same  period,  ArcelorMittal  HR  sales  shares  increased  by […]p.p. The analysis also estimates that the sales shares of Imports and of Other EEA suppliers have decreased by […]p.p. and […]p.p., respectively. However, those coefficients are not statistically significant (i.e. they are not statistically different from zero). (27) Thus, the OLS regression analysis at Southern European level indicates that the sales shares loss of Ilva following the 2012 Ilva Cap were […], while  Imports and Other EEA suppliers did not increase their sales shares in the same period.

(70)                       Regarding the 2015 Ilva Cap and the dumping period (ilvacap2015_dumping), as expected Imports gained significant sales shares in Southern Europe with an increase in HR sales shares of […]p.p. At the same time, Ilva HR sales shares decreased       by […]p.p., followed by Other EEA suppliers that lost  […]p.p.  ArcelorMittal  gained […]p.p. HR sales shares, although the coefficient is not statistically significant. As explained in paragraph (51), those coefficients capture the joint effect on HR sales shares of both the imposition of the 2015 Ilva Cap and the dumped volumes from imports, and hence their interpretation in terms of competitive interaction across market participants are not clear, in particular for the coefficients of the Ilva and ArcelorMittal regressions. This is because ArcelorMittal might have gained sales shares from the 2015 Ilva Cap […] and, simultaneously, lost sales shares from the increased of imports due to the dumping. Similarly, the coefficient  estimated for the Ilva regression captures both the structural loss of sales shares following the reduction in its capacity and production, and the loss due to  the dumped imported volumes.

(71)                       In relation to the AD measures (AD), the analysis' results are somewhat unclear. Post-AD  measures,  Imports  gained  […]p.p.  sales  shares  while  ArcelorMittal  lost […]p.p. sales shares. This is somewhat surprising, and might be related to the fact that the dataset only covers a limited number of months following the imposition of the AD measures. From most recent data on imports, it appears that from the second half of 2017 HR imported volumes started decreasing sharply, as discussed in Section 9.4.5.7 of the Decision.

(72)                      Turning to the coefficients of the HR market prices in Southern Europe and China, the Imports regression shows that for a 1% increase in Southern  European  HR market prices, […]p.p. of sales shares would be diverted from domestic EEA suppliers to imports. Likewise, for a 1% increase in Chinese HR market prices, […]p.p. of sales shares would be diverted from imports to domestic EEA suppliers. The fact that the two coefficients are very similar in absolute value indicates the relevance of the relative HR market price between Southern Europe and China in the availability of imports in Southern Europe, as suggested in Section 3.

(73)                      The coefficients of the domestic market prices further shows that imports cannot be expected to have a disproportionate volume reaction to a change in HR market price in Southern Europe. The coefficient of the Southern European HR domestic market price indicates that if HR market prices in Southern Europe increased by 5%, the HR sales share of imports would increase by approximately […]p.p. Thus, in a hypothetical post-Transaction scenario, the reaction from imports does not appear to be as sizeable as to defeat the incentives of the merger entity to increase prices. Furthermore, it should be noted that the Southern European prices in the analysis are average market prices. Thus, the merged entity would need to increase its prices significantly more in order to reach a market wide price increase of 5%.

(74)                     The coefficients of the domestic Southern European HR market price in the regressions of the other market participants have the expected negative sign, indicating that their sales shares decrease as prices in Southern Europe increase to the benefit of imports.

(75)                    Overall, the results of the OLS regressions in Southern Europe indicate that ArcelorMittal and Ilva are close competitors in the supply of HR, and that imports  are unlikely to exert a sufficient constraint to the merged entity post-Transaction as  to defeat its incentives to increase prices.

(76)                   Table 3 presents the results for the regressions for Italy, which broadly confirm the conclusions abovementioned.

(77)                      The 2012 Ilva Cap caused a decrease in HR sales shares of […]p.p. for Ilva in Italy, a steeper decrease in sales shares compared to Southern Europe. This is not surprising, as Italy is the country where Ilva's HR sales are mostly concentrated. In the same period ArcelorMittal gained […]p.p. HR sales shares, whereas Other European Suppliers gained […]p.p. (with a coefficient significant at […]% level). The volume losses of Ilva were therefore distributed among ArcelorMittal and Other EEA Suppliers, with ArcelorMittal gaining […] of it. As in the Southern European regressions, the coefficient in the Imports regression is negative and not statistically significant.

(78)                     Following the 2015 Ilva Cap and the dumping period,  imports  gained  approximately […]p.p. HR sales shares in Italy. That is a greater gain compared to the Southern Europe regressions. Again, this is not surprising as Italy is the EEA country that absorbs the greatest imported HR volumes. Similarly to the regressions for Southern Europe, though, Ilva lost […]p.p. HR sales shares in Italy, while Other EEA Suppliers lost […]p.p. The only noteworthy change between Southern Europe and Italy is the coefficient in ArcelorMittal regression that, although not statistically significant in both sets of regressions, passes from positive to negative.

(79)                     Contrary to the Southern European regression, in the regressions for Italy the coefficient of the AD dummy has the expected sign. Imports decreased by approximately […]p.p. following the AD measures (with a coefficient significant at […]% level), while the other EEA based suppliers gained, although the coefficients are not statistically significant.

(80)                    Regarding the coefficients of domestic HR prices, the analysis estimates that in Italy imports gain approximately […]p.p. in HR sales shares from domestic EEA suppliers for a given 1% increase in HR domestic market Southern European prices, and lose about […]p.p. in HR sales shares for a given 1% increase in domestic Chinese HR prices. Compared to the results in Southern Europe, the magnitude of the coefficients is higher, but the two coefficients are still very close. Furthermore, also in Italy the Other EEA Suppliers appear to be particularly affected by domestic prices in Southern Europe, to the benefit of imports.

6.2. Results: IV Regressions

(81)                   This Section presents the results of the same regressions using instrumental variable estimation. The instruments used by the Commission are the raw material prices used for the production of steel, specifically the iron ore price and the scrap metal price.

(82)                  Table 4 presents the results of the IV regression for Southern Europe. The coefficients are very similar to the ones in the OLS regression. In particular, the coefficients of ilvacap2012 still indicate a […] sales shares gain from ArcelorMittal following the sales shares reduction of Ilva, and the coefficient of the HR Southern European price in the Import  regression  still  indicate  an  increase  of  approximately […]p.p. of imports in Southern Europe following an increase by 1% of the HR Southern European prices.

(83)                   The panel at the bottom of the table reports the p-values of several statistical tests used in IV regressions to gather insights on the validity of the instruments. Overall, the results of the statistical tests provide reassurance that the instruments are valid and are not weak.

(84)                  Table 5 displays the results of the IV regression for Italy. As in the case of Southern Europe, the coefficients are very similar to the OLS regression. About […] of the sales shares lost by Ilva ([…]p.p.) following the 2012 Ilva Cap were gained by ArcelorMittal ([…]p.p.), whereas Imports sales shares remained constant and Other European Suppliers gained the remainder ([…]p.p.), although the latter coefficient is not statistically significant.

(85)                 The coefficient of the Southern European HR price in the Import regression  is slightly higher compared to the OLS regression, and is approximately […]p.p., suggesting that an 5% increase in Southern European HR prices would lead  to imports gaining about […]p.p. sales shares in Italy. Again, such imports reaction  does not appear to be such as to defeat the incentives to increase prices by the  merged entity.

7. ROBUSTNESS TESTS

(86)                    This Section presents the robustness tests of the Commission's quantitative analysis. The robustness tests are meant to check whether and to what extent the coefficients  in the baseline scenarios discussed in Section 6 are subject to changes by varying some assumptions or parameters of the model. Hence, they test whether the conclusions drawn from the Commission analysis' baseline scenarios hold true under different assumptions.

(87)                   The Commission presents three robustness tests. All robustness tests are run for Southern Europe with IV estimation. In Section 7.1 the period analysed is limited until 2015. In Section 7.2 the period analysed starts from January 2011 onwards. In Section 7.3 the Commission uses only one instrument in a specification called "just identified". Finally, Section 7.4 presents further robustness tests included  in  the Letter of Facts following the comments of the Notifying Party included in its Reply  to the SO.

7.1. Robustness test: Period until 2015

(88)                  As discussed in Section 3, it appears that the high correlation between imported volumes to Europe and the relative domestic HR price in Southern Europe and in China becomes lower from 2015 onwards. Possibly, this is due to the dumping  period in EEA starting in 2015. In any event, by limiting the period under analysis to the period before 2015, the robustness test presented in this Section investigates whether the coefficient on the Southern European HR price in the Import regression is affected by the period of low correlation between relative price and imported volumes.

(89)                Table 6 shows that the Commission's conclusions do not change substantially by excluding data on the period from January 2015 onwards. The coefficient on the Southern European HR price in the Import regression increases only slightly to […], indicating that a 1% increase in Southern European price would lead to an increase in imported volume by […]p.p.

(90)               The coefficients on the 2012 Ilva Cap are essentially unchanged, while the coefficients on 2015 Ilva Cap – dumping period and on the post-AD measures are missing because the period covered by the two dummy variables has been excluded from the analysis.

7.2.  Robustness test: Period from 2011

(91)                       The second robustness test investigates whether limiting the period under analysis to a more recent period has any effect on the analysis. Specifically, this robustness test limits the analysis to the period from 2011 onwards. This excludes the economic crisis in 2008/9 and its aftermath, in case the control variables included in  the analysis to control for demand shifts (i.e. GDP and worldwide consumption)  were not able to fully account for that global demand shock.

(92)                      The first effect of this robustness is that the endogeneity test discussed in Section 5.1 finds indication of endogeneity. (31) Therefore, the OLS results for this specification would be biased, whereas the IV results provide consistent estimates.

(93)                      As shown in Table 7, there are also changes in some coefficients. In particular, the coefficient on the Southern European HR price in the Import regression, although still positive, is no longer statistically significant. Furthermore, the p-value of the coefficient is high, which indicates that the probability that the coefficient is different from zero is low. This result indicates that, for the period from 2011 onwards, the  HR sales shares of imports in Southern Europe were not related to the Southern European HR price. Thus, even if Southern European prices increased, the imports sales shares would not increase.

(94)                    As regards the coefficients on the dummy variables, they are broadly in line with the coefficients estimated in the baseline scenario in Table 4, and so are the conclusions.

7.3. Robustness test: Just Identified Specification

(95)                     An IV estimation with one endogenous variable and one instrument is called just identified. The Commission uses a just identified specification of its baseline IV estimation for Southern Europe to test the robustness of the instruments. If the results do not change substantially in the just identified specification, it means that the results are relatively robust to the choice of the instruments.

(96)                   In Table 8 the Commission uses as instrument only the scrap metal price. The coefficients on the dummy variables are in line with the baseline scenario of Table 4, and so are the conclusions. The coefficient on the Southern European price in the Import specification is lower than the baseline scenario at […], indicating that a 1% increase in Southern European prices leads to an increase in imports HR sales shares by […]p.p. Hence, in the just identified specification the volume reaction of imports to Southern European prices is reduced. This supports the findings in the Commission's baseline specification, and suggests that the imports' reaction to Southern European HR prices might be even lower.

7.4. Further Robustness tests included in the Letter of Facts

(97)                      The Commission carried out further robustness tests to the quantitative analysis to test whether the Commission's conclusions would be affected by the criticisms of the Notifying Party included in its Reply the SO (Annex 3.B.1). In particular, those robustness tests focus on the elasticity estimation of the sales share of domestic suppliers, that is the coefficient on the Southern European HR price variable.

(98)                     The regressions are run with the instrumental variables estimation, with the dependent variable being the aggregate sales share of Domestic EEA Suppliers (ssh_DOMEEA_HR), (32) rather than separately for ArcelorMittal, Ilva and Other EEA Suppliers. As explained in paragraph (67), the resulting coefficient on the price variable will be equal to the coefficient of the Imports regression (not reported in the following tables) with opposite sign. As above, the resulting coefficient represents  the percentage points change to the sales share of domestic EEA suppliers that is diverted to imports.

(99)                    Moreover, the instruments (the prices of the raw materials iron ore and scrap metal) are now expressed in logarithms.

(100)                  The following specifications are included:

(a)   Column 1: Exclusion of the dummy variable identifying the Ilva capacity cap of 2012 (ilvacap2012), and introduction of the Industrial Price Index (in logarithm) as control variable (lIPI);

(b)   Column 2: Exclusion of the first difference of GDP (D.gdp) as control variable;

(c)   Column 3: Exclusion of the first difference of worldwide consumption (D.ww_consumption) as control variable;

(d)   Column 4: Same variables as Column 1, but including the dummy variable for the 2012 Ilva Cap among the instruments (in third lag).

(101)                   The aim of those specifications is to test whether the results of the Commission's quantitative analysis significantly change with the exclusion of the 2012 Ilva Cap, that could be considered as a supply shifter, and the further inclusion and exclusion of demand shifters (Industrial Price Index, GDP and worldwide consumption).

(102)                   As in the previous analyses, the regressions include exogenous factors affecting the availability and prices of imports (namely, the domestic price in China and the dummy variables identifying the dumping period and the post–anti-dumping measures), and hence the extent to which customers can switch from EEA domestic producers to imports following an increase in the Southern European domestic price.

(103)                  Furthermore, the results are presented at three levels of data aggregation, the same levels used by the Notifying Party in its Reply to the SO. Namely, (i) disaggregated Southern Europe including Italy, Spain, Portugal and Greece (the same level of data aggregation used in the previous results); (ii) disaggregated Southern Europe including Italy, Spain, and Portugal (i.e. excluding Greece, as the Notifying Party pointed out that the results are susceptible to its exclusion); and (iii) aggregated Southern Europe.

(104)                  Table 9 presents the results for Southern Europe disaggregated into Italy, Spain, Portugal and Greece; Table 10 presents the results for Southern Europe  disaggregated into Italy, Spain and Portugal (i.e. excluding Greece); Table 11 presents the results for the aggregated Southern Europe.

(105)                   Furthermore, Table 12 to Table 14 present the results for the same levels of aggregation of Southern Europe as the previous tables, but restricting the period of the analysis to 2011-2017H1 (i.e. excluding from the sample the economic crisis of 2008/09 and its immediate aftermath).

(106)                  Table 9 to Table 14 show that the coefficient estimate for the Southern European price, when statistically significant, ranges between […] and […], indicating that  a 1% increase in Southern European prices, all else being equal, is estimated to decrease the combined sales shares of EEA domestic producers by a range of […] and […]p.p. Consequently, the imports sales shares would increase by the same amount. Based on those estimates, for a 5% increase in the Southern European  market price, the imports sales shares would increase by approximately […]p.p.

(107)                   The Commission also estimated the residual demand elasticity directly, through an estimation of (log) volume sales. The analysis over the full period, however, does not provide meaningful residual demand elasticity estimates (i.e. the elasticity estimates have a positive sign). Estimating elasticity from time series is often challenging and unstable likely due to greater difficulties in separating time trends from causal relations between price and volumes. However, when the analysis is restricted to the shorter period from  2011  onwards  (which  notably  excludes  the  economic  crisis of 2009), the elasticity estimates become more stable and have the expected sign.

(108)                  Table 15 to Table 17 replicate Table 12 to Table 14 using log volume sales of domestic   EEA   producers   as   dependent   variable   (lvol_DOMEEA_HR). The interpretation of the coefficient of Southern European price is the elasticity of the residual demand.

(109)              The estimates of the elasticity of the residual demand range between […] and […]. Taking into account only the estimates from Table 16 (excluding Greece) and Table 17 (Southern Europe aggregated), and the specification where all demand shifters are present (columns 1 and 4), the estimated elasticity of the residual demand ranges between […] and […].

(110)              The elasticity of the residual demand for domestic suppliers is the sum of the aggregate demand elasticity and the elasticity of the share of domestic producers. (33) Based on the Commission's estimates presented above on the share reaction to price changes (no higher than […]), the elasticity of the share of domestic suppliers is no higher than […]. (34)

(111)              Combining those estimates implies an aggregate elasticity of demand between approximately […] and […] (subtracting the elasticity of the share of domestic producers from the residual demand elasticity). This range for the aggregate  elasticity of demand appears reasonable and is consistent with the estimates of the aggregate elasticity of demand in the U.S. provided by the U.S. International Trade Commission for HR in 2011, ranging between -0.75 and -1. (35),(36) This provides further confidence on the robustness of the results of the Commission's quantitative analysis. Furthermore, it shows that the aggregate demand for HR products in Southern  Europe appears to be relatively inelastic.

(112)              In its Reply to the Letter of Facts (Annex 1), the Notifying Party carried out a regression specification similar to the Commission's specifications, but replacing the domestic Chinese price with the imports price in Southern Europe. The Notifying Party found higher residual demand elasticities, albeit none of the coefficient was statistically significant. This is not surprising. The Notifying Party's specifications suffer from multi-collinearity (37), as the domestic Southern European price and price  of imports in Southern Europe are highly correlated. This multi-collinearity problem inflates the magnitude of the coefficients of both variables, but renders them not statistically significant (i.e. the regression is not able to disentangle the effects of the two variables). Moreover, contrary to domestic Chinese price, the imports price in Southern Europe is likely endogenous. However, the raw material prices used as instrumental variables are not appropriate to be instruments of both domestic and imports' prices in Southern Europe, as they have the same explanatory power on both price variables. This results in weak instruments, and further inflates the coefficients. Therefore,  the  Commission's  considers  that the  Notifying  Party's  approach is not appropriate to estimate the residual demand of EEA suppliers. (38), (39)

(113)              Overall, the results of the further checks presented above support the conclusion reached by the Commission that the increase in the share of imports is unlikely to be sufficient to defeat the post-merger price increase by the merged entity.

8. CONCLUSION

(114)              The Commission carried out a regression analysis investigating the closeness of competition among competitors in the HR market by analysing the sales shares gains following the imposition of the Ilva Cap. Additionally, the Commission's quantitative analysis provided insights regarding the diversion of sales shares from domestic EEA suppliers to imports following a HR price increase in Southern Europe and Italy.

(115)             The Commission has further tested the robustness of its analysis by (i) excluding the period as of January 2015 from the analysis, (ii) reducing the period of the analysis from   January    2011    onwards,    (iii)    using    only    one    instrumental variable, (iv)       introducing new control variables and changing the level of aggregation of the data, in response to the comments made by the Notifying Party.

(116)           Overall, the results of the Commission's quantitative analysis lead to the following conclusions.

(117)           First, there are consistent indications through the specifications that following the 2012 Ilva Cap, ArcelorMittal gained […] the sales shares loss by Ilva. As for the period from 2015, the results are unclear as two simultaneous events (the 2015 Ilva Cap and the dumping period) prevent from drawing inferences on the estimated coefficients as regards the closeness of competition between the Parties, although imports have gained share significantly and Ilva has lost share significantly during that period. As noted in Section 9.4.2.7 of the Decision, however, in light of the qualitative evidence gathered during the market investigation, the Commission cannot exclude that during the dumping period, the diversion of Ilva sales also benefitted imports.

(118)          Second, the volumes diverted from domestic EEA suppliers to imports following     an increase in Southern European HR price appears to be insufficient as to defeat the incentives to increase prices by the merged entity. The quantitative analysis estimates that for a 1% increase in Southern European HR prices imports would increase their sales shares by at most […]p.p. in Southern Europe, and […]p.p. in Italy. This means that for a 5% increase in Southern European HR prices, imports' sales shares would increase by approximately […]p.p. in Southern Europe, and by […]p.p. in Italy. It should be recalled that the analysis examines market prices in Southern Europe and  in Italy. Therefore, the merged entity should increase its prices by even more to reach a market wide price increase of 5%. Accordingly, the imports' HR sales shares reactions in Southern Europe and in Italy estimated by the Commission's analysis do not appear to be sufficient as to defeat the merged entity's incentives to increase price post-merger.

(119)             In conclusion, the Commission considers that its quantitative analysis corroborates the qualitative and quantitative evidence gathered during the market investigation, as it indicates that the Parties are close competitors, and that imports, although exercising a competitive constraint, would not be able to constrain the merged entity's incentives to increase prices. Furthermore, it should be noted that the Commission's estimates as regards the imports' reaction do not fully account for the AD measures recently imposed on several importing countries. As such, the ability and the incentives of imports to constrain the merged entity may be even lower than estimated in the Commission's analysis as set out in this Annex.

Annex 2

April 11, 2018

Case M.8444 – ArcelorMittal / Ilva

COMMITMENTS TO THE EUROPEAN COMMISSION

Pursuant to Article 8(2) of Council Regulation (EC) No 139/2004 (the “Merger Regulation”), ArcelorMittal S.A. (“ArcelorMittal”) hereby enters into the following Commitments (the “Commitments”) vis-à-vis the European Commission (the “Commission”) with a view to rendering its acquisition of sole control over certain Italian steel business units of the Ilva Group (“Ilva” and, together with ArcelorMittal, the “Parties”) (the “Transaction”) 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 Transaction 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”).

Section A. Definitions

1.               For the purpose of the Commitments, the following terms shall have the following meaning:

Affiliated Undertakings: undertakings controlled by ArcelorMittal, 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”).

AM Invest: AM InvestCo Italy S.r.l., incorporated under the laws of the Italian Republic, with its registered office at viale Brenta 27/28, I-20139 Milan, Italy.

ArcelorMittal: ArcelorMittal S.A., incorporated under the laws of the Grand Duchy of Luxembourg, with its registered office at boulevard d’Avranches, 24-26, L-1160 Luxembourg.

Assets: the assets that are necessary to ensure the viability and competitiveness of the Divestment Businesses as indicated in Section B, paragraph 5(a), (b), (c), and (d), and described more in detail in the Schedules.

Closing: the transfer of the legal title to the Divestment Businesses to the Purchaser(s).

Closing Period: each period of 3 months from the approval of the Purchaser and the terms of sale by the Commission in relation to a Divestment Business.

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 Businesses: the businesses defined in Section B and in the Schedules which ArcelorMittal commits to divest.

Divestment Commitment: the commitment to divest the Divestment Business described in Section B.

Divestiture Trustee: one or more natural or legal person(s) who is/are approved by the Commission and appointed by ArcelorMittal and who has/have received from ArcelorMittal the exclusive mandate to sell the Divestment Businesses at no minimum price.

Effective Date: the date of adoption of the Decision.

First Divestiture Period: the period of […] from the Effective Date.

Hold Separate Managers: the persons appointed by ArcelorMittal for the Divestment Businesses to manage the day-to-day business under the supervision of the Monitoring Trustee, i.e. […].

Key Personnel: all personnel necessary to maintain the viability and competitiveness of the Divestment Businesses, as listed in the Schedules, including the Hold Separate Managers.

Liège/Dudelange Business: the Divestment Business described more in detail in Schedule 2.

Marcegaglia: Marcegaglia Steel S.p.A., incorporated under the laws of Italy, with registered office at Via Bresciani 16, 46040 Gazoldo degli Ippoliti, Mantova, Italy.

Marcegaglia Commitment: the commitment to cause Marcegaglia to divest its shareholdings in the AM Invest consortium and for ArcelorMittal to disengage from its obligation to acquire a shareholding in Marcegaglia Business, described in Section C.

Monitoring Trustee: one or more natural or legal person(s) who is/are approved by the Commission and appointed by ArcelorMittal, and who has/have the duty to monitor ArcelorMittal’s compliance with the conditions and obligations attached to the Decision.

Ostrava Business: the Divestment Business described more in detail in Schedule 3.

Parties: ArcelorMittal and Ilva.

Personnel: all staff currently employed by the Divestment Businesses.

Piombino/Galati Business: the Divestment Business described more in detail in Schedule 1.

Purchaser(s): one or more entities approved by the Commission as the acquirer(s) of the Divestment Businesses in accordance with the criteria set out in Section E.

Purchaser Criteria: the criteria laid down in paragraph Section E of these Commitments that the Purchaser(s) must fulfil in order to be approved by the Commission.

Schedules: the schedules to these Commitments describing more in detail the Divestment Businesses.

Trustee(s): the Monitoring Trustee and/or the Divestiture Trustee as the case may be.

Trustee Divestiture Period: the period of […] from the end of the First Divestiture Period.

Section B. The Divestment Commitment

Commitment to divest

2.                  In order to maintain effective competition, ArcelorMittal commits to divest, or procure the divestiture of the Divestment Businesses by the end of the Trustee Divestiture Period as going concern(s) to the Purchaser(s) and on terms of sale approved by the Commission in accordance with the procedure described in paragraph 17 of these Commitments. To carry out the divestiture, ArcelorMittal commits to find one or more Purchasers and to enter into final binding sale and purchase agreement(s) for the sale of the Divestment Businesses by the end of the First Divestiture Period. If ArcelorMittal has not entered  into such an agreement by the end of the First Divestiture Period in relation to any Divestment Business, ArcelorMittal shall grant the Divestiture Trustee an exclusive mandate to sell that Divestment Business in accordance with the procedure described in paragraph 30 in the Trustee Divestiture Period.

3.                   ArcelorMittal shall be deemed to have complied with the Divestment Commitment if:

(a)    By the end of the Trustee Divestiture Period, ArcelorMittal or the Divestiture Trustee has entered into final binding sale and purchase agreement(s) and the Commission approves the proposed Purchaser(s) and the terms of sale as being consistent with the Commitments in accordance with the procedure described  in paragraph 17.

(b)    The Closing of the sale(s) of the Divestment Businesses to the Purchaser(s) takes place within the Closing Period(s).

4.                  In order to maintain the structural effect of the Commitments, ArcelorMittal 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 Businesses, unless, following the submission of a reasoned request from ArcelorMittal showing good cause and accompanied by a report from the Monitoring Trustee (as provided in paragraph 44 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 Businesses is no longer necessary to render the Transaction compatible with the internal market.

Structure and definition of the Divestment Businesses

5.                The Divestment Businesses comprise:

(a)    The Piombino/Galati Business described in Schedule 1.

(b)    The Liège/Dudelange Business described in Schedule 2.

(c)    The Ostrava Business described in Schedule 3.

The legal and functional structure of the Divestment Businesses as operated to date is described in the Schedules. The Divestment Businesses, described in more detail in the Schedules, include all assets and staff that are necessary to ensure the viability and competitiveness of the Divestment Businesses, in particular:

(a)      All tangible and intangible assets (including the intellectual property rights specified in the Schedules).

(b)      All    licences,   permits   and   authorisations   issued   by   any governmental organisation for the benefit of the Divestment Businesses.

(c)      All contracts, leases, commitments, and customer orders (order books) of the Divestment Businesses, and all customer, credit and other records of the Divestment Business.

(d)     The Personnel.

Section C. The Marcegaglia Commitment

6.                  ArcelorMittal shall cause Marcegaglia to divest its shareholdings in the AM Invest consortium. ArcelorMittal shall also disengage from its obligation to acquire a shareholding in Marcegaglia, […].

7.                   ArcelorMittal shall be deemed to have complied with the Marcegaglia Commitment […].

Section D. Related Commitments

Preservation of viability, marketability and competitiveness

8.                  From the Effective Date until Closing, ArcelorMittal shall preserve or procure the preservation of the economic viability, marketability and competitiveness of the Divestment Businesses, in accordance with good business practice, and shall minimise as far as possible any risk of loss of competitive potential of the Divestment Businesses. In particular, ArcelorMittal undertakes:

(a)    Not to carry out any action that might have a significant adverse impact on the value, management or competitiveness of the Divestment Businesses or that might alter the nature and scope of activity, or the industrial or commercial strategy or the investment policy of the Divestment Businesses.

(b)    To make available, or procure to make available, sufficient resources for the development of the Divestment Businesses, on the basis and continuation of  the existing business plans.

(c)    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 Businesses, and  not to solicit or move any Personnel to ArcelorMittal’s remaining businesses. Where, nevertheless, individual members of the Key Personnel exceptionally leave the Divestment Businesses, ArcelorMittal shall provide a reasoned proposal to replace the person or persons concerned to the Commission and the Monitoring Trustee. ArcelorMittal 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.

Hold-separate obligations

9.                   From the Effective Date until Closing, ArcelorMittal shall keep the Divestment Businesses separate from the businesses it is retaining and ensure that, unless explicitly permitted under these Commitments: (i) management and staff of the business retained  by ArcelorMittal have no involvement in the Divestment Businesses; (ii) the Key Personnel and Personnel of the Divestment Businesses have no involvement in any business retained by ArcelorMittal and do not report to any individual outside the Divestment Businesses.

10.                Until Closing, ArcelorMittal shall assist the Monitoring Trustee in ensuring that the Divestment Businesses are managed as distinct and saleable entities separate from the businesses which ArcelorMittal is retaining. Immediately after the adoption of the Decision, ArcelorMittal shall appoint the Hold Separate Managers. The Hold Separate Managers, who shall be part of the Key Personnel, shall manage the Divestment Businesses  independently  and  in  the  best  interest  of  the  businesses  with  a  view  to ensuring their continued economic viability, marketability and competitiveness and their independence from the businesses retained by ArcelorMittal. The Hold Separate Managers shall closely cooperate with and report to the Monitoring Trustee and, if applicable, the Divestiture Trustee. Any replacement of the Hold Separate Managers  shall be subject to the procedure laid down in paragraph 8(c) of these Commitments. The Commission may, after having heard ArcelorMittal, require ArcelorMittal to replace one or more Hold Separate Manager(s).

Ring-fencing

11.                  ArcelorMittal shall implement, or procure to implement, all necessary measures to ensure that it does not, after the Effective Date, obtain any Confidential Information relating to the Divestment Businesses and that any such Confidential Information obtained by ArcelorMittal before the Effective Date will be eliminated and not be used by ArcelorMittal. This includes measures vis-à-vis ArcelorMittal’s appointees on the supervisory boards and/or boards of directors of the Divestment Businesses.  In  particular, the participation of the Divestment Businesses in any central information technology network shall be severed to the extent possible, without compromising the viability of the Divestment Businesses. ArcelorMittal may obtain or keep information relating to the Divestment Businesses which is reasonably necessary for the divestiture of the Divestment Businesses or the disclosure of which to ArcelorMittal is required by law.

Non-solicitation clause

12.                  ArcelorMittal undertakes, subject to customary limitations, not to solicit, and to procure that Affiliated Undertakings do not solicit, the Key Personnel transferred with the Divestment Businesses for a period of […] after Closing.

Due diligence

13.                  In order to enable potential purchasers to carry out a reasonable due diligence of the Divestment Businesses, ArcelorMittal 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 Businesses.

(b)    Provide to potential purchasers sufficient information relating to the Personnel  and allow them reasonable access to the Personnel.

Reporting

14.                  ArcelorMittal shall submit written reports in English on potential purchasers of the Divestment Businesses 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). ArcelorMittal shall submit a list of all potential purchasers having expressed interest in acquiring the Divestment Businesses 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.

15.                  ArcelorMittal 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.

Section E. The Purchaser(s)

16.                  In order to be approved by the Commission, the Purchaser(s) must fulfil the following criteria:

(a)    The Purchaser(s) shall be independent of and unconnected to ArcelorMittal and its Affiliated Undertakings (this being assessed having regard to the situation following the divestiture).

(b)    The Purchaser(s) shall have the financial resources, proven expertise, and incentive to maintain and develop the Divestment Businesses as viable and active competitive forces in competition with ArcelorMittal and other competitors.

(c)    The acquisition of the Divestment Businesses by the Purchaser(s) 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 Businesses.

(d)       […].

(e)       […].

17.                  ArcelorMittal will organize an open sales process and will keep the Commission and the Monitoring Trustee regularly informed about its progress, to enable the Monitoring Trustee to ensure that all potentially suitable purchasers are given the opportunity to participate in the sales process on an equal and non-discriminatory footing.

18.                  The final binding sale and purchase agreement(s) (as well as ancillary agreements) relating to the divestment of the Divestment Businesses shall be conditional on the Commission’s approval. When ArcelorMittal 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. ArcelorMittal must be able to demonstrate to the Commission that the Purchaser(s) fulfil(s) the Purchaser Criteria and that the Divestment Businesses are being sold in a manner consistent with the Commission’s Decision and the Commitments. For the approval, the Commission shall verify that the Purchaser(s) fulfil(s) the Purchaser  Criteria and that the Divestment Businesses are being sold in a manner consistent with the Commitments including their objective to bring about a lasting structural change in the market.  The Commission may approve the sale of a 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 Businesses after the sale, taking account of the proposed Purchaser(s).

Section F. Trustee

I.    Appointment procedure

19.                  ArcelorMittal shall appoint a Monitoring Trustee to carry out the functions specified in these Commitments for a Monitoring Trustee. ArcelorMittal commits not to close the Transaction before the appointment of a Monitoring Trustee.

20.                  If ArcelorMittal has not entered into binding sale and purchase agreement(s) regarding any of the Divestment Businesses one month before the end of the First Divestiture  Period or if the Commission has rejected the Purchaser(s) proposed by ArcelorMittal at that time or thereafter, ArcelorMittal shall appoint a Divestiture Trustee for the Divestment Business(es) in relation to which ArcelorMittal has not entered into a final and binding sale and purchase agreement or in relation to which the Commission has rejected the Purchaser(s) proposed by ArcelorMittal. The appointment of the Divestiture Trustee shall take effect upon the commencement of the Trustee Divestiture Period.

21.                  The Monitoring Trustee shall:

(a)    At the time of appointment, be independent of ArcelorMittal and its Affiliated Undertakings.

(b)    Possess the necessary qualifications to carry out its mandate, for example have sufficient relevant experience as an investment banker or consultant or auditor.

(c)    Neither have nor become exposed to a Conflict of Interest.

22.                  The Trustee shall be remunerated by ArcelorMittal in a way that does not impede the independent and effective fulfilment of its mandate. In particular, where  the  remuneration package of a Divestiture Trustee includes a success premium linked to the final sale value of a Divestment Business, such success premium may only be earned if the divestiture takes place within the Trustee Divestiture Period.

Proposal by ArcelorMittal

23.                  No later than two weeks after the Effective Date, ArcelorMittal shall submit the name or names of one or more natural or legal persons whom ArcelorMittal proposes to appoint as the Monitoring Trustee to the Commission for approval. No later than one month before the end of the First Divestiture Period or on request by the Commission, ArcelorMittal shall submit a list of one or more persons whom ArcelorMittal 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 fulfil the requirements set out in paragraph 21 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.

(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

24.                  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, ArcelorMittal 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, ArcelorMittal 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 ArcelorMittal

25.                  If all the proposed Trustees are rejected, ArcelorMittal 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 23 and 24 of these Commitments.

Trustee nominated by the Commission

26.                  If all further proposed Trustees are rejected by the Commission, the Commission shall nominate a Trustee, whom ArcelorMittal shall appoint, or cause to be appointed, in accordance with a trustee mandate approved by the Commission.

II.  Functions of the Trustee

27.                  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 ArcelorMittal, 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

28.                  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 Managers, the on-going management of the Divestment Businesses with a view to ensuring their continued economic viability, marketability and competitiveness and monitor compliance by ArcelorMittal 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 Businesses, and the keeping separate of the Divestment Businesses from the business retained by ArcelorMittal, in accordance with paragraphs 8 and 9 of these Commitments.

b)            Supervise the management of the Divestment Businesses as distinct and saleable entities, in accordance with paragraph 10 of these Commitments.

c)             With respect to Confidential Information:

-                 Determine all necessary measures to ensure that ArcelorMittal does not after the Effective Date, obtain any Confidential Information relating to the Divestment Businesses.

-                 In particular strive for the severing of the Divestment Businesses’ participation in a central information  technology network to the extent possible, without compromising the viability of the Divestment Businesses.

-                 Make sure that any Confidential Information relating to the Divestment Businesses obtained by ArcelorMittal before the Effective Date is eliminated and will not be used by ArcelorMittal.

-                 Decide whether such information may be disclosed to or kept by ArcelorMittal as the disclosure is reasonably necessary to allow ArcelorMittal to carry out the divestiture or as the disclosure is required by law.

d)         Monitor the splitting of assets and the allocation of Personnel between the Divestment Businesses and ArcelorMittal or Affiliated Undertakings.

iii.           Propose to ArcelorMittal such measures as the Monitoring Trustee considers necessary to ensure ArcelorMittal’s 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 Businesses, the holding separate of the Divestment Businesses 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 Businesses and the Personnel in particular by reviewing, if available, the data room documentation, the information memorandum and the due diligence process.

b)       Potential purchasers are granted reasonable access to the Personnel.

v.            Act as a contact point for any requests by third parties in relation to the Commitments.

vi.          Provide to the Commission, sending ArcelorMittal 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 Businesses 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.

vii.         Promptly report in writing to the Commission, sending ArcelorMittal a non-confidential copy at the same time, if it concludes on reasonable grounds that ArcelorMittal is failing to comply with these Commitments.

viii.        Within one week after receipt of a documented proposal referred to in paragraph 17 of these Commitments, submit to the Commission, sending ArcelorMittal a non-confidential copy at the same time, a reasoned opinion as to the suitability and independence of the proposed Purchaser(s) and the viability of the Divestment Business after the sale and as to whether the Divestment Businesses are sold in a manner consistent with the conditions and obligations attached to the Decision, in particular, if relevant, whether the sale of the Divestment Businesses without one or more Assets or not all of the Personnel affects the viability of the Divestment Businesses 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.

29.                  If the Monitoring and Divestiture Trustee are not the same legal or natural 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

30.                  Within the Trustee Divestiture Period, the Divestiture Trustee shall sell at no minimum price the Divestment Business(es) in relation to which ArcelorMittal has not entered into a final and binding sale and purchase agreement or in relation to which the Commission has rejected the Purchaser(s) proposed by ArcelorMittal, to one or more Purchasers, provided that the Commission has approved both the Purchaser(s) and the final binding sale and purchase agreement(s) (and ancillary agreements) as in line with the Commission’s Decision and the Commitments in accordance with Section E and paragraph 17 of these Commitments. The Divestiture Trustee shall include in the sale  and purchase agreement(s) (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 ArcelorMittal, subject to ArcelorMittal’s unconditional obligation to divest at no minimum price in the Trustee Divestiture Period.

31.                  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 ArcelorMittal.

III.  Duties and obligations of ArcelorMittal

32.                  ArcelorMittal 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 any of ArcelorMittal’s or the Divestment Businesses’ books, records, documents, management or other personnel, facilities, sites and technical information necessary for fulfilling its duties under the Commitments and ArcelorMittal and the Divestment Businesses shall provide the Trustee upon request with copies of any document. ArcelorMittal and the Divestment Businesses 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.

33.                  ArcelorMittal shall provide the Monitoring Trustee with all managerial and  administrative support that it may reasonably request on behalf of the management of  the Divestment Businesses. This shall include all administrative support functions relating to the Divestment Businesses which are currently carried out at headquarters level. ArcelorMittal 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. ArcelorMittal 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.

34.                  ArcelorMittal 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(s) and the Closing, including the appointment of advisors to assist with the sale process. Upon request of the Divestiture Trustee, ArcelorMittal shall cause the documents required for effecting the sale(s) and the Closing to be duly executed.

35.                  ArcelorMittal 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 ArcelorMittal 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 wilful default, recklessness, gross  negligence or bad faith of the Trustee, its employees, agents or advisors.

36.                  At the expense of ArcelorMittal, the Trustee may appoint advisors (in particular for corporate finance or legal advice), subject to ArcelorMittal’s 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 ArcelorMittal refuse to approve the advisors proposed by the Trustee the Commission may approve the appointment of such advisors instead, after having heard ArcelorMittal. Only the Trustee shall be entitled to issue instructions to the advisors. Paragraph 35 of these Commitments shall apply mutatis mutandis. In the Trustee Divestiture Period, the Divestiture Trustee may use advisors who served ArcelorMittal during the Divestiture Period if the Divestiture Trustee considers this in the best interest of an expedient sale.

37.                  ArcelorMittal agrees that the Commission may share Confidential Information  proprietary to ArcelorMittal 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.

38.                  ArcelorMittal agrees 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 of the identity and the tasks of the Monitoring Trustee.

39.                  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

40.                  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 ArcelorMittal, require ArcelorMittal to replace the Trustee; or

(b)    ArcelorMittal may, with the prior approval of the Commission, replace the Trustee.

41.                  If the Trustee is removed according to paragraph 40 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 19-26 of these Commitments.

42.                  Unless removed according to paragraph 40 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.

Section G. The Review Clause

43.                  The Commission may extend the time periods foreseen in the Commitments in response to a request from ArcelorMittal or, in appropriate cases, on its own initiative. Where ArcelorMittal requests an extension of a time period, it shall submit a reasoned request to the Commission no later than one 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 ArcelorMittal. Only in exceptional circumstances shall ArcelorMittal be entitled to request an  extension within the last month of any period.

44.                  The Commission may further, in response to a reasoned request from ArcelorMittal 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 ArcelorMittal. 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.

Section H. Entry Into Force

45.                  The Commitments shall take effect upon the date of adoption of the Decision.

(signed)

SCHEDULE 1 – THE PIOMBINO/GALATI BUSINESS […]

SCHEDULE 1(A) – THE PIOMBINO PLANT […]

SCHEDULE 1(B) – THE GALATI PLANT […]

SCHEDULE 1(C) – THE SKOPJE PLANT […]

SCHEDULE 2 – THE LIÈGE/DUDELANGE BUSINESS […]

SCHEDULE 2(A) – THE LIÈGE PLANT […]

SCHEDULE 2(B) – THE DUDELANGE PLANT […]

SCHEDULE 3 – THE OSTRAVA BUSINESS […]

SCHEDULE 3 – THE OSTRAVA PLANT […]

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                  OJ C ...,...200. , p....

3                  OJ C ...,...200. , p....

4                  Publication in the Official Journal of the European Union No C 329, 30.9.2017, p. 14.

5                  […].

6                  […].

7                  […].

8                  Turnovers relate to 2016.

9                  Commission Regulation (EC) No 802/2004 of 7 April 2004 implementing Council Regulation (EC) No 139/2004 on the control of concentrations between undertakings (OJ L 133, 30.4.2004, p. 1).

10                 In  this  respect  the  Notifying  Party  cites  Case  T-31/99,  ABB  Asea  Brown  Boveri  v. Commission, judgment of 20 March 2002, paragraph 99.

11                 Ibid.

12                 For instance, […], Annex 1.1 to the response to RFI 20.

13                 See Annex 1.8 to reply to RFI 20.

14                 See Annex 1.3 to reply to RFI 20.

15                 Doc ID1469-32173 [internal document].

16                 Response to the Statement of Objections, p. 189.

17                 Response to Letter of Facts, p. 3.

18                 This is in contrast to, for instance, stainless steel that contains a considerable amount of chromium.

19                 It is also technically possible to cool the liquid hot metal into pig iron and use it later to feed a steel- making process. However, such a process involves an additional production step and also results in a loss of energy (heat). In the integrated route the hot metal is typically kept in liquid state and fed immediately to a BOF.

20                 This contrasts with long products of which, according to the Notifying Party, approximately  […]% are produced through the EAF route.

21                 Reply to question 56 of RFI 7, 10 September 2017.

22                 Reply to question 56 of RFI 7, 10 September 2017.

23                 See,   for   instance,   […]   dated 8.6.2016,   as   presented   in   the   Letter   of   Facts,    Section   1.11; Doc ID1660-54969 [internal document].

24                 Re-rollers may however be structurally dependent on the HR producers that can offer them the requisite security of supply. [Commercial information].

25                 See recital (186).

26                 Replies to question 56.1 of Q3 – Questionnaire to Customers (Industrial); Doc ID2800.

27                 Replies to question 56.1 of Q3 – Questionnaire to Customers (Industrial); Doc ID2800.

28                 Replies to question 56.1 of Q4 – Questionnaire to Customers (SSCs); Doc ID2804 (machine translation from Italian).

29                 Replies to question 56.1 of Q4 – Questionnaire to Customers (SSCs); Doc ID2804 (machine translation from Italian).

30                 Doc ID1658-1160 [internal document].

31                 ArcelorMittal is a public limited liability company (société anonyme) incorporated under the laws of the Grand Duchy of Luxembourg. ArcelorMittal is a holding company with no business operations of  its own. All of ArcelorMittal's significant operating subsidiaries are indirectly owned by ArcelorMittal through intermediate holding companies. Source: Annual Report 2016 p. 5 and p. 30.

32                 As of 31 December 2016, 37.4% of the ordinary shares are owned by the Significant Shareholder, which is a trust of which Mr. Lakshmi N. Mittal and Mrs. Usha Mittal and their children are the beneficiaries. Source: ArcelorMittal Annual Report 2016 p. 152.

33                 Ownership as of 31 December 2016, based on ArcelorMittal Annual report 2016.

34                 Ownership as of 31 December 2015, based on ArcelorMittal Annual report 2016.

35                 If we take into account the production capacities available at the Taranto site. Though currently,  due to environmental constrains, the production lines are not allowed to produce at those capacity levels.

36                 Commercial sites of the Ilva group in Greece and Tunisia are not part of the Transaction.

37                 Source: Slide 3 of Annex 23.1 of reply to RFI 19 – Project Trullo Valuation presentation. The HR capacity reported in the figure diverges from the more accurate capacity figures provided by the Notifying Party in the later stages of the proceedings.

38                 This share would amount to […]% under the hypothesis that Ilva could fully utilize its available HR capacity.

39                 ArcelorMittal internal document, Doc ID2431-465, [internal document].

40                 ArcelorMittal internal document, reply to RFI 19, slide 15 of Annex 23.1.

41                 Production sites located in Verona, Caronno Petrusella (Varese), Lesegno (Cuneo),    Malegno, Sellero, Cerveno (Brescia), and Annone Brianza (Lecco).

42                 All the production assets are owned by Ilva S.p.A. standalone (without subsidiaries.

43                 Converted with amendments by Law no. 231 of 24 December 2012. (‘Law Decree no. 207/2012’).

44                Based on Law Decree no. 61/2013 of 4 June 2013, converted into Law no. 89/2013 on 3   August 2013. Law no. 89/2013 was further amended on 10 December 2013 by Article 7 of Law Decree no. 136/2013, converted into Law no. 6/2014 on 6 February 2014.

45                 See Official Journal no. 105 of 8 May 2014, available at: http://www.gazzettaufficiale.it/eli/id/2014/05/08/14A03637/sg.

46                 Compiled based on the following sources: […].

47                 […].

48                 Confirmed non-confidential minutes of a call with a competitor on 17.7.2017, Doc ID268.

49                 RFI 25 Annexes – […].

50                 For the avoidance of doubt, Ilva S.p.A. is representative of Ilva Group.

51                 ArcelorMittal document, Doc ID2431-465, […].

52                 […].

53                 This ad hoc procedure was modelled on the so-called Marzano law, a special insolvency law organising the extraordinary administration of large undertakings in difficulty that are entrusted with essential public services.

54                 Doc ID3772; http://www.gruppoilva.com/it/press-release/2017-05-26/comunicato-stampa.

55                 Doc ID3771;         http://www.sviluppoeconomico.gov.it/index.php/it/194-comunicati-stampa/2036649- calenda-firma-il-decreto-di-aggiudicazione-del-complesso-industriale-del-gruppo-ilva-ad-am-investco- italy.

56                 Doc ID3771;         http://www.sviluppoeconomico.gov.it/index.php/it/194-comunicati-stampa/2036649- calenda-firma-il-decreto-di-aggiudicazione-del-complesso-industriale-del-gruppo-ilva-ad-am-investco- italy.

57                 Form CO, paragraph 555.

58                 Form CO, section 7 III B.

59                 Form CO, paragraph 556.

60                 Form CO, section 7 III B.

61                 Form CO, paragraphs 563–4.

62                 Confirmed non-confidential minutes of a call with a market participant on 12.7.2017, Doc ID323.

63                 Form CO, paragraph 557.

64                 Form CO, section 7 III B.

65                 Form CO, section 7 III B.

66                 Form CO, paragraph 559.

67                 Form CO, section 7 III B.

68                 Form CO, paragraph 561.

69                 Form CO, section 7 III B.

70                 Form CO, paragraph 560.

71                 Form CO, paragraphs 565 and 567.

72                 Reply to RFI 2 Marcegaglia, on AD 630 non-confidential, Comments on threat of injury, causal link and Union interest p. 12/40.

73                 The   total   pro-forma   consolidated   sales   revenue   of   Marcegaglia   Steel   S.p.A.   in    2015   was EUR 3.84 billion, (out of which Italy represents EUR 3.4 billion). Pro-forma financials were prepared  to reflect Profit & Loss figures as if Marcegaglia Steel S.p.A. had been operating from 1 January 2015. Source: Reply to RFI 2 Marcegaglia non-confidential version.

74                 In Italy the  locations are  Gazaldo,  Albignasego,  Boltiere,  Brenta,  Caslmaggiore,  Contino, Corsico, Dusino San Michele, Forli, Lainate, Lomagna, Montechiarugolo, Nogaro, Osteria Grande, Pozzolo Formigaro and Ravenna.

75                 See, for instance […]. Doc ID1658-35043 [internal document].

76                 Reply to question 16 of RFI 1.

77                 DWT: Deadweight tonnage is a measure of a vessel's weight carrying capacity, and does not include the weight of the ship itself.

78                 Confirmed non-confidential minutes of a call with a market participant on 12.7.2017, Doc ID323.

79                 Confirmed non-confidential minutes of a call with a market participant on 12.7.2017, Doc ID323.

80                 Confirmed non-confidential minutes of a call with a market participant on 12.7.2017, Doc ID323.

81                 Confirmed non-confidential minutes of a call with a market participant on 6.12.2017, Doc ID3686.

82                 Confirmed non-confidential minutes of a call with a market participant on 12.7.2017, Doc ID323.

83                 Reply to RFI 2 Marcegaglia on AD measures non-confidential version.

84                 Form CO.

85                 Reply to question 1of RFI Marcegaglia, 22 September 2017.

86                 Confirmed non-confidential minutes of a call with a market participant on 6.12.2017, Doc ID3686.

87                 Reply to question 1 of RFI Marcegaglia, 22 September 2017.

88                 Reply to question 1 of RFI Marcegaglia, 22 September 2017.

89                 Source: Commission’s computation based on Eurostat trade data.

90                 Imports  data  from  Eurostat  are  provided  aggregated  at  the  extra  EU28  level.  For  this reason the considered flow of imports is from outside the EU28 to the EU28 countries. The same analysis conducted at the extra EEA level would not materially change the results.

91                 Regulation  (EU)  2016/1036  of  the  European  Parliament  and  of  the  Council  of  8  June   2016  on protection against dumped imports from countries not members of the European Union (codification) (OJ L 176, 30.6.2016, p. 21).

92                 Article 5 of Regulation (EU) 2016/1036.

93                 Article 6(9) of Regulation (EU) 2016/1036.

94                 Article 7(1) of Regulation (EU) 2016/1036.

95                 Article 11 of Regulation (EU) 2016/1036.

96                 Commission Implementing Regulation (EU) 2016/1778 of 6 October 2016 imposing a provisional anti- dumping duty on imports of certain hot-rolled flat products of iron, non-alloy or other alloy steel originating in the People's Republic of China (OJ L 272, 7.10.2016, p. 33).

97                 Commission  Implementing  Regulation (EU)  2017/649  of 5  April  2017  imposing  a definitive anti- dumping duty on imports of certain hot-rolled flat products of iron, non-alloy or other alloy steel originating in the People's Republic of China (OJ L 92, 6.4.2017, p. 68).

98                 Commission   Implementing   Regulation   (EU)   2017/969   of   8   June   2017    imposing   definitive countervailing duties on imports of certain hot-rolled flat products of iron, non-alloy or other alloy steel originating in the People's Republic of China and amending Commission Implementing Regulation (EU) 2017/649 imposing a definitive anti-dumping duty on imports of certain hot-rolled flat products of iron, non-alloy or other alloy steel originating in the People's Republic of China (OJ L 146, 9.6.2017,   p. 17).

99                 Commission Implementing Regulation (EU) 2017/5 of 5 January 2017 making imports of   certain hot- rolled flat products of iron, non-alloy or other alloy steel originating in Russia and Brazil subject to registration (OJ L 3, 6.1.2017, p. 1).

100               Commission Implementing Regulation (EU) 2017/1795 of 5 October 2017 imposing a    definitive anti- dumping duty on imports of certain hot-rolled flat products of iron, non-alloy or other alloy steel originating in Brazil, Iran, Russia and Ukraine and terminating the investigation on imports of certain hot-rolled flat products of iron, non-alloy or other alloy steel originating in Serbia (OJ L 258, 6.10.2017, p. 24).

101               Commission  Implementing  Regulation  (EU)  2015/2325  of  11  December  2015  making imports of certain cold-rolled flat steel products originating in the People's Republic of China and the Russian Federation subject to registration (OJ L 328, 12.12.2015, p. 104).

102               Commission Implementing  Regulation (EU)  2016/181 of 10  February 2016  imposing   a provisional anti-dumping duty on imports of certain cold-rolled flat steel products originating in the People's Republic of China and the Russian Federation (OJ L 37, 12.2.2016, p. 1).

103               Commission Implementing Regulation (EU) 2016/1328 of 29 July 2016 imposing a definitive anti- dumping duty and collecting definitively the provisional duty imposed on imports of certain cold rolled flat steel products originating in the People's Republic of China and the Russian Federation (OJ L 210, 4.8.2016, p. 1).

104               Commission Implementing  Regulation (EU)  2016/1329  of 29  July 2016  levying the  definitive anti- dumping duty on the registered imports of certain cold-rolled flat steel products originating in the People's Republic of China and the Russian Federation (OJ L 210, 4.8.2016, p. 27).

105               Commission  Implementing  Regulation  (EU)  2017/1238  of  7  July  2017  making  imports of certain corrosion resistant steels originating in the People's Republic of China subject to registration (OJ L 177, 8.7.201, p. 39).

106               Commission Implementing Regulation (EU) 2017/1444 of 9 August 2017 imposing a  provisional anti- dumping duty on imports of certain corrosion resistant steels originating in the People's Republic of China (OJ L 207, 10.8.2017, p. 1).

107               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).

108               […].

109               Replies to question 95 of Q1 – Questionnaire to competitors (EEA suppliers of flat carbon steels), Doc ID2798, and Confirmed non-confidential minutes of  a  call  with  a  competitor  on  18.7.2017,  Doc ID434.

110               See recital (155).

111               See recital (88).

112               Source:    Worldsteel,    steel    demand    crude    steel    equivalent;    -    RoW:    Rest    of the   world. https://www.worldsteel.org/.../17.11.2016_Adam_Szewczyk_world+steel+outlook.pdf.

113               M.8444ID002283-000003, […], p. 11.

114               USB Global I/O: Global Steel Supply and Demand, 7 April 2016, p. 28.

115               See,     for     instance,     M.4137     –      Mittal/Arcelor,     paragraphs     14     and      15;               M.6974  –  Metinvest/Lanebrook/Southern Gok, paragraphs 27–9.

116               See, for instance, M.7155 – SSAB/Rautaruukki; M.6471 – Outokumpu/Inoxum, M.4137 – Mittal/Arcelorand ECSC.1351 – Usinor/Arbed/Aceralia.

117               See,  for  instance,  M.7155 – SSAB/Rautaruukki, paragraphs 23–5 and  41; M.4137    – Mittal/Arcelor, paragraphs 17–8; and ECSC.1351 – Usinor/Arbed/Aceralia, paragraphs 32–3.

118               See,  for  instance,  M.7155 – SSAB/Rautaruukki, paragraphs 24–5 and  41; M.4137    – Mittal/Arcelor, paragraphs 18–37; and ECSC.1351 – Usinor/Arbed/Aceralia, paragraphs 34–67.

119               See,  for   instance,  M.7155  –   SSAB/Rautaruukki;  M.4137   –  Mittal/Arcelor;   and   ECSC.1351 – Usinor/Arbed/Aceralia.

120               See, for instance, M.7155 – SSAB/Rautaruukki¸ paragraphs 24–5 and 41; M.4137 – Mittal/Arcelor, paragraph 22; and ECSC.1351 – Usinor/Arbed/Aceralia, paragraphs 42–4.

121               Replies to question 7 of Q3 – Questionnaire to customers (industrial) Doc ID2800, 814, 1347, 1466, 1872, 3555, 3620 and 3621; and replies to question 7  of  Q4  –  Questionnaire to  customers (SSCs) Doc ID2804.

122               Replies to question 7 of Q3 – Questionnaire to customers (industrial) Doc ID2800, 814, 1347, 1466, 1872, 3555, 3620 and 3621; and replies to question 7  of  Q4  –  Questionnaire to  customers (SSCs)  Doc ID2804; replies to questions 74–6 of Q11 –  Questionnaire to  Industrial  Customers (Phase  II)  Doc ID3225, 3624 and 3625.

123               ECSC.1237 – Arbed/Aceralia, paragraph 18, ECSC.1243 – KruppHoesch/Thyssen, paragraph    21, and ECSC.1268 – Usinor/Cockerill Sambre, paragraph 18.

124               M.4408 – Tata/Corus, paragraphs 13–7, and M.4515 – CSN/Corus, paragraphs 10–3 and 15.

125               ECSC.1351 – Usinor/Arbed/Aceralia, paragraphs 58–62.

126               M.4137 – Mittal/Arcelor, paragraph 31, and M.4992 – ArcelorMittal/Galvex, paragraph 11.

127               See,   for   instance,   M.5771   –   CSN/Cimpor,   paragraph   19,   M.7461   –    AMDS   Italia/CLN/JV, paragraphs 26–8,  M.7762 – ArcelorMittal/Financial  Entities/Grupo  Condesa,  paragraphs  16–8,  and M.8159 – ArcelorMittal/Cellino/JV, paragraphs 15–7. See also Voestalpine/Profilafroid/SAP, where the Commission ultimately left the exact market definition open but appears to have assessed the case at the level of GS; M.4291 – Voestalpine/Profilafroid/SAP, paragraphs 13, 15 and 38. See also SSAB/Rautaruukki where the Commission noted that the Parties were not equally active in all segments of GS, such as EG; M.7155 – SSAB/Rautaruukki, paragraphs 24–5 and 68.

128               Form CO, Figure 45.

129               ArcelorMittal’s reply to the Letter of Facts, paragraph 32.

130               See,  for  instance,  Commission  Notice  on  the  definition  of  relevant  market  for  the    purposes  of Community competition law, OJ C 372, 9.12.1997, p. 5 (‘Commission Notice on relevant markets, paragraph 2.

131               See, for instance, Commission Notice on relevant markets, paragraphs 7 and 13.

132               See, for instance, Commission Notice on relevant markets paragraphs 20 and 23.

133               Replies to  questions 12  and  13  of Q1 – Questionnaire to  Competitors (EEA suppliers of   flat carbon steel), Doc ID2798.

134               Replies to  questions 12  and  13  of Q1 – Questionnaire to  Competitors (EEA suppliers of   flat carbon steel). Doc ID2798.

135               Replies to question 16 of Q1 – Questionnaire to Competitors (EEA suppliers of flat carbon steel), Doc ID2798.

136               Replies to question 36 of Q12 – Questionnaire to Steel distributors / SSCs (Phase II), Doc ID3223.

137               Replies to question 36 of Q11 – Questionnaire to Industrial Customers (Phase II), Doc ID3225, 3624 and 3625.

138               Replies to question 44 of Q11 – Questionnaire to Industrial Customers (Phase II), Doc ID3225, 3624 and 3625.

139               Replies to question 43 of Q11 – Questionnaire to Industrial Customers (Phase II), Doc ID3225, 3624 and 3625.

140               Replies to questions 43 and 44 of Q11 – Questionnaire to Industrial Customers (Phase II), Doc ID3225, 3624 and 3625.

141               Replies to question 42 of Q11 – Questionnaire to Industrial Customers (Phase II), Doc ID3225, 3624 and 3625 and replies to question 35 of Q12 – Questionnaire to Steel distributors / SSCs (Phase II),    Doc ID3223.

142               ArcelorMittal, response to the Letter of Facts.

143               ArcelorMittal, response to the Letter of Facts.

144               See,   for   example,   M.7155   –   SSAB/Rautaruukki,   paragraph   96,   and   M.7061      –   Huntsman Corporation/Equity interests held by Rockwood Holdings, recital 146 and Annex I.

145               Replies  to  questions  33  and  34  of  Q12  –  Questionnaire  to  Steel  distributors  /  SSCs  (Phase  II), Doc ID3223, and replies to questions 40 and 41 of Q11 – Questionnaire to  Industrial  Customers  (Phase II), Doc ID3225, 3624 and 3625.

146               Replies to question 18 of Q13 – Questionnaire to Competitors (Phase II), Doc ID3224.

147               Replies to questions 40 and 41 of Q11 – Questionnaire to Industrial Customers (Phase II), Doc ID3225, 3624 and 3625 and replies to questions 8 and 9 of Q13 – Questionnaire to Competitors (Phase II),     Doc ID3224.

148               Replies to question 41 of Q11 – Questionnaire to Industrial Customers (Phase II), Doc ID3225, 3624 and 3625.

149               Form CO, Table 8.

150               Information in brackets has been calculated by the Commission on the basis of figures provided  by the Notifying Party. […].

151               Form CO, paragraph 163.

152               ECSC.1351 – Usinor/Arbed/Aceralia, paragraphs 58–62.

153               Doc ID1658-71506 [internal document].

154               […].

155               […].

156               See,   for   instance,   M.4137  –   Mittal/Arcelor,   paragraph  37;   and   M.7155  –  SSAB/Rautaruukki, paragraph 24.

157               See, for instance, M.4408 – Tata Steel/Corus; M.4515 – CSN/Corus; M.4137 – Mittal/Arcelor; M.5771– CSN/CIMPOR.

158               Form CO, paragraph 285.

159               Form CO, paragraph 286.

160               Form CO, paragraph 286; and Response to the Article 6(1)(c) decision, paragraph 222.

161               Form CO, paragraph 286; and Response to the Article 6(1)(c) decision, paragraph 222.

162               Replies to question 18 of Q1 – Questionnaire to Competitors (EEA Suppliers of flat carbon steels); Doc ID2798.

163               Replies to question 21 of Q1 – Questionnaire to Competitors (EEA Suppliers of flat carbon steels); Doc ID2798.

164               Replies to questions 19 and 19.1 of Q3 – Questionnaire toCustomers (industrial); Doc ID2800.

165               Replies to question 20.1 of Q3 – Questionnaire to Customers (industrial); Doc ID2800.

166               Replies to questions 20, 21 and 21.1 of Q3 – Questionnaire to Industrial Customers; Doc ID2800.

167               Confirmed non-confidential minutes of a call with a customer on 12.6.2017; Doc ID1975.

168               Confirmed non-confidential minutes of a call with a customer on 12.12.2017; Doc ID3741.

169               Replies to questions 24 and 24.1 of Q3 – Questionnaire to Customers (industrial); Doc ID2800.

170               Replies to question 25 of Q3 – Questionnaire to Customers (industrial); Doc ID2800.

171               Replies to question 25.1 of Q3 – Questionnaire to Customers (industrial); Doc ID2800;  and Confirmed non-confidential minutes of a call with a customer on 12.12.2017, Doc ID3741.

172               See,    for    example,    Case    M.0315    –    Mannesmann/Vallourec/Ilva,    Commission  decision   of 31 January 1994,    M.0484    –    Krupp/Thyssen/Riva/Tadfin/AST,     Commission     decision     of     21 December 1994, M.2382 – Usinor/Arbed/Aceralia, Commission decision of 21 November 2001 and M.7762 – ArcelorMittal/Financial Entities/Grupo Condesa, Commission decision of 29 January 2016.

173               M.2382   –   Usinor/Arbed/Aceralia,   Commission   decision   of   21   November   2001,     M.1203   – Usinor/Finarvedi, Commission decision of 29 September 1998, M.1014 – British Steel/Europipe, Commission decision of 26 February 1998, M.906 – Mannesmann/Vallourec, Commission decision     of 3 June 1997.

174               M.0484  –  Krupp/Thyssen/Riva/Tadfin/AST,  Commission  decision  of  21  December  1994;  M.906 –Mannesmann/Vallourec, Commission decision of 3 June 1997.

175               M.7762 – ArcelorMittal/Financial Entities/Grupo Condesa, Commission decision of 29 January 2016.

176               M.2382   –   Usinor/Arbed/Acerali,   Commission   decision   of   21   November   2001,       M.1203   – Usinor/Finarvedi, Commission decision of 29 September 1998. See also, for example M.1595 – British Steel/Hoogovens, Commission decision of 15 July 1999; M.906 – Mannesmann/Vallourec, Commission decision  of  3  June  1997.  M.484  –  Krupp/Thyssen/Riva/Falck/Tadfin/AST,  Commission  decision of 21 December    1994;    and     M.315      –     Mannesmann/Vallourec/Ilva,    Commission                           decision of 31 January 1994.

177               M.7762 – ArcelorMittal/Financial Entities/Grupo Condesa, Commission decision of 29 January 2016.

178               Replies to Question 7 of Q6 - Questionnaire to Tube Customers. Doc ID2806. Replied to Question  8 of Questionnaire to Tube Producers.

179               Ibid. paragraph 27.

180               Replies to Question 9 and10 of Q6 – Questionnaire to Tube Customers. Doc ID2806. Replies to Question 10 of Q5 – Questionnaire to Tube Producers, Doc ID2085.

181               Replies  to  Question  6  (RFI  to  Tube  Customers)  and  Question  5  (RFI  to  Tube  Producers), sent 26/9/2017.

182               Replies to Question 11 of Q5 - Questionnaire to Tube Producers, Doc ID2085

183               COMP/IV/M.1014    –    British    Steel/Europipe,    Commission    decision    of    26  February   1998, paragraphs 25-32.

184               Ibid, paragraphs 28-29.

185               See, for instance, M.7461 – AMDS Italia/CLN/JV, paragraphs 19 and 21; M.7155 – SSAB/Rautaruukki, paragraph 48; and M.4137 – Mittal/Arcelor, paragraph 80.

186               See, for instance M.4137 – Mittal/Arcelor; M.6974 – Metinvest/Lanebrook/Southern GOK; and M.7762– ArcelorMittal/Financial Entities/Grupo Condesa.

187               M.7155 – SSAB/Rautaruukki, paragraph 40.

188               Confirmed non-confidential minutes of a call with a competitor on 18.7.2017, Doc ID434.

189               Confirmed non-confidential minutes of a call with a competitor on 6.4.2017, Doc ID2100.

190               See, for instance, replies to questions 74 and 75 of Q1 – Questionnaire to Competitors   (EEA suppliers of flat carbon steels), Doc ID2798.

191               See, for instance, replies to question 76 of Q1 – Questionnaire to Competitors (EEA suppliers of flat carbon steels), Doc ID2798.

192               See, for example, confirmed non-confidential minutes of calls with competitors, Doc ID443, 1776 as well as response from a competitor to question 76 of Q1 – Questionnaire to Competitors (EEA suppliers of flat carbon steels).

193               Confirmed  non-confidential  minutes  of  calls  with  competitors  on  6.4.17,  Doc  ID2100,      and  on 17.7.2017, Doc ID268 and 1776.

194               See, for instance, confirmed non-confidential minutes of a call with a competitor, Doc ID443.

195               See,  for  instance,  M.7155  –  SSAB/Rautaruukki;  M.6974  –     Metinvest/Lanebrook/Southern  GOK; M.1310 – British Steel/Hoogovens; M.3326 LNM/PHS; and M.4137 – Mittal/Arcelor.

196               See, for instance, M.4515 CSN/Corus; and M.4408 – Tata Steel/Corus.

197               Form CO, p. 97.

198               See,     for     instance,     ECSC.1237     –     Arbed/Aceralia,     paragraphs     21–4;      ECSC.1243     – KruppHoesch/Thyssen, paragraphs 26, 27 and 29; ECSC.1351 – Usinor/Arved/Aceralia, paragraphs 88, 89, 91, 94 and 96; M.4137 – Mittal/Arcelor, paragraph 67; and M.4992 – ArcelorMittal/Galvex, paragraph 12.

199               Response to the Article 6(1)(c) decision, paragraph 20.

200               Response to the Article 6(1)(c) decision, Figure 6.

201               See,  for  instance,  Commission  Notice  on  the  definition  of  relevant  market  for  the    purposes  of Community competition law, OJ C 372, 9.12.1997, p. 5 (‘Commission Notice on Relevant Markets’), paragraph 8.

202               Commission Notice on Relevant Markets, paragraphs 44–52.

203               Replies to questions 33 and 62–7 of Q3 – Questionnaire to customers (industrial), Doc ID2800, 814, 1347, 1466, 1872, 3555, 3620 and 3621; replies to questions 33 and 62–7 of Q4 – Questionnaire to

customers (SSCs), Doc ID2804; replies to questions 19, 22, 88 and 89 of Q11 – Questionnaire to Industrial Customers (Phase II), Doc ID3225, 3624 and 3625; replies to questions 16, 19, 76 and 77 of Q12 – Questionnaire to Steel distributors / SSCs (Phase II), Doc ID3223.

204               ArcelorMittal’s reply to RFI 23 Annex 6.1 (Prices) and Annex 6.3 (raw material costs).

205               Doc ID4252.

206               […].

207               Horizontal  Merger  Guidelines,  paragraph  28. Horizontal Merger  Guidelines,  footnote 32: "Products may be differentiated in various ways. There may, for example, be differentiation in terms of  geographic location, based on branch or stores location."

208               See, for instance, M.7878 –HeidelbergCement/Schwenk/Cemex Hungary/Cemex Croatia.

209               See also Section 5.3.1.

210               Annex 29.1 to the Notifying Party’s reply to the Commission’s RFI 19.

211               Sales to the whole of France.

212               Replies  to  questions  11,  48  and  83  of  Q11  –  Questionnaire  to  Industrial  Customers   (Phase  II), Doc ID3225, 3624 and 3625; and replies to questions 11, 40 and 71 of Q12 – Questionnaire to Steel distributors / SSCs (Phase II), Doc ID3225, 3624, 3625.

213               Confirmed   non-confidential   minutes   of   a   call   with   an   anonymous   market       participant   on 15 December 2017, Doc ID3574.

214               Confirmed non-confidential minutes of a call with a customer on 25 April 2017, Doc ID1946.

215               Confirmed non-confidential minutes of a call with a customer on 26 April 2017, Doc ID3631.

216               Confirmed  non-confidential  minutes  of  a  meeting  between  the  Commission's  case  team      and  a competitor on 21 November 2017, Doc ID3183.

217               Replies to questions 14–6 of Questionnaire 13 to Competitors (Phase II), Doc ID3224. See  also replies to question 39 of Questionnaire 1 to Competitors (EEA suppliers of flat carbon steels), Doc ID2798 and confirmed non-confidential  minutes  of  conference  calls  with  market  participants  on  25.4.2017  Doc ID102, 26.4.2017 Doc ID278 and 12.7.2017 Doc ID2062.

218               Replies to questions 14–6 of Q13 – Questionnaire to Competitors (Phase II), Doc ID3224.

219               Reply to question 39 of Q1 – Questionnaire to Competitors (EEA suppliers of flat carbon steels), Doc ID2798.

220               Dated 25 November 2016 with the subject line 'FW: Price Meteo W47' forwarding the 'price   meteo' of that week.

221               Doc ID1469-54302 [internal document].

222               Replies  to  questions  11,  48  and  83  of  Q11  –  Questionnaire  to  Industrial  Customers   (Phase  II), Doc ID3225, 3624 and 3625; replies to questions 11, 48 and 71 of Q12 - Questionnaire to Steel distributors / SSCs (Phase II), Doc ID3223; and replies to questions 14–6 of Q13 – Questionnaire to Competitors (Phase II), Doc ID3224. See also, for instance, confirmed non-confidential minutes of a  call with a customer on 26.4.2017, Doc ID278 and confirmed non-confidential minutes of a call with a customer on 7.12.2017, Doc ID3761.

223               Commission's computation on the basis of sales information provided by the Notifying Party. Sales for 2016 are computed as follows. For AM, Ilva, Marcegaglia, imports and market size: actual data provided by the Parties. For competitors the Parties only provided  sales  data  up  to  September/October 2016. Sales in the missing months have been then attributed to competitors on the basis of their sales during the period January/September 2016.

224               See, for instance, confirmed non-confidential minutes of a call with a steel producer on 17.7.2017, Doc ID268,  and  confirmed  non-confidential  minutes  a   call   with  a   distributor   on  12.6.2017,  Doc ID1744.  See  also  replies  to  question  32  of  Q3 –  Questionnaire  to  customers  (industrial),  Doc ID2800, 814, 1347, 1466, 1872, 3555, 3620 and 3621.

225               See, for instance, confirmed non-confidential minutes of a call with a steel producer on 17.7.2017, Doc ID1776, confirmed non-confidential minutes  of  a  call  with  a  steel  producer  on  18.7.2017,  Doc ID443 and confirmed minutes of a call with a steel producer 25.7.2017, Doc ID2938. See also

replies to questions 39 and 44 of Q1 – Questionnaire to Competitors (EEA suppliers of flat carbon steels), Doc ID2798.

226               Capacities provided by the Notifying Party.

227               Eurostat Comext database and data provided by ArcelorMittal (response to RFI 19, Annex 29.1).

228               France indicates the whole of France and is not split between Southern and Northern France.

229               See, for instance, replies to questions 39 and 44 of Q1 – Questionnaire to competitors (EEA suppliers of flat carbon steels), Doc ID2798. See also confirmed non-confidential minutes of a call with a  competitor on 18.7.2017, Doc ID443.

230               Replies to question 44 of Q1 – Questionnaire to Competitors (EEA suppliers of flat carbon steels), Doc ID3798.

231               Replies to question 32 of Q3 – Questionnaire to customers (industrial), Doc ID2800.

232               See, for instance, replies to question 36 of Q3 – Questionnaire to customers (industrial), Doc ID2800, 814, 1347, 1466, 1872, 3555, 3620 and 3621; and replies to question 36 of Q4 – Questionnaire to customers (SSCs), Doc ID2804.

233               Commission’s computation based on data provided by Ilva.

234               Commission’s computation based on data provided by Ilva.

235               Commission’s computation based on data provided by Ilva

236               Commission’s computation based on data provided by ArcelorMittal (response to RFI 23, Annex 1). Actual sales for the period January-October 2016, projections for the period November-December 2016.

237               Commission’s computation based on data provided by ArcelorMittal (response to RFI 23, Annex 1). Actual sales for the period January-October 2016, projections for the period November-December 2016.

238               Commission’s computation based on data provided by ArcelorMittal (response to RFI 23, Annex 1). Actual sales for the period January-October 2016, projections for the period November-December 2016.

239               Commission’s computation based on data provided by ArcelorMittal (response to RFI 23, Annex 1). Actual sales for the period January-October 2016, projections for the period November-December 2016.

240               Commission’s computation based on data provided by ArcelorMittal (response to RFI 23, Annex 1). Actual sales for the period January-October 2016, projections for the period November-December 2016.

241               Commission’s computation based on data provided by ArcelorMittal (response to RFI 23, Annex 1). Actual sales for the period January-October 2016, projections for the period November-December 2016.

242               Commission’s computation based on data provided by ArcelorMittal.

243               Doc ID1469-6380 [internal document].

244               Commission’s computation based on data provided by ArcelorMittal.

245               Commission’s computation based on data provided by ArcelorMittal.

246               Replies to questions 35–37 of Q1 – Questionnaire to competitors (EEA suppliers of flat  carbon steels), Doc ID2798. See also confirmed non-confidential minutes of a call with a competitor on 17.7.2017,  Doc ID268.

247               Confirmed non-confidential minutes of a call with a competitor on 17.7.2017, Doc ID268.

248               Eurostat Comext database and data provided by ArcelorMittal (response to RFI 19, Annex 29.1).

249               Replies to question 30 of Q4 – Questionnaire to customers (SSCs), Doc ID2804.

250               Replies to question 30 of Q3 – Questionnaire to customers (industrial), Doc ID2800, 814,   1347, 1466, 1872, 3555, 3620 and 3621.

251               Confirmed non-confidential minutes of a call with an industrial customer on 26.4.2017, Doc ID278.

252               Confirmed non-confidential minutes of a call with a market participant on 12.7.2017, Doc ID323.

253               Confirmed non-confidential minutes of a call with a distributor customer on 12.6.2017, Doc ID1744.

254               See for instance Form CO, paragraph 530.

255               See, for instance, replies to questions 39 and 40 of Q1 – Questionnaire to competitors (EEA suppliers of flat carbon steels), Doc ID2798; replies to question 32 of Q3 – Questionnaire to customers (industrial), Doc ID2800, 814, 1347, 1466, 1872, 3555, 3620 and 3621; and replies to question 32 of Q4 – Questionnaire to customers (SSCs), Doc ID2804. See also replies to question 11 of Q11 –  Questionnaire to Industrial Customers (Phase II), Doc ID3225, 3624 and 3625, replies to question 11 of Q12 – Questionnaire to Steel distributors / SSCs (Phase II), Doc ID3223, and replies to question 14 of Q13 – Questionnaire to Competitors (Phase II), Doc ID3224. See also confirmed non-confidential minutes of calls with competitors on 12.7.2017 Doc ID465, 17.7.2017 Doc ID1776 and 18.7.2017     Doc ID443.

256               See, for instance, replies to questions 15 and 16 of Q13 – Questionnaire to Competitors (Phase II), Doc ID3224, replies to questions 48 and 83 of Q11 – Questionnaire to Industrial Customers (Phase II), Doc ID3225, 3624 and 3625, and replies to questions 40 and 71 of Q12 – Questionnaire to Steel distributors / SSCs (Phase II), Doc ID3223.

257               Contrary to the Notifying Party's argument, the difference between prices in Northern Europe and prices in Southern Europe cannot be solely explained by a difference in transport costs of non-EEA exporters. From data provided by the Notifying Party (Reply to RFI 26, Annex 7), the difference in freight to  reach Southern Europe (Genoa, Italy) and Northern Europe (Antwerp, Belgium) is between […] from India and China. Moreover, exporting countries such as Russia and Ukraine have an even lower freight reaching Antwerp than reaching Genoa.

258               SBB, Notifying Party's Reply to RFI 23, Annex 6.1.

259               SBB, Notifying Party's Reply to RFI 23, Annex 6.1.

260               SBB, Notifying Party's Reply to RFI 23, Annex 6.1.

261               Commission's  calculations  on  the  basis  of  ArcelorMittal  reply  to  RFI  26  (prices)  and      RFI  23 (Annex 6.3).

262               Commission's  calculations  on  the  basis  of  ArcelorMittal  reply  to  RFI  26  (prices)  and      RFI  23 (Annex 6.3).

263               Commission's  calculations  on  the  basis  of  ArcelorMittal  reply  to  RFI  26  (prices)  and      RFI  23 (Annex 6.3).

264               See,   for   example,   M.7155   –   SSAB/Rautaruukki,   paragraph   96,   and   M.7061      –   Huntsman Corporation/Equity interests held by Rockwood Holdings, recital 146 and Annex I.

265               Notifying  Party's  Response  to  the  Article  6  (1)(c)  decision,  Annex  1.1,  page  42 (the 'conditional correlation').

266               Price series for other Southern European countries, such as Greece or Portugal, was not available.

267               Confirmed non-confidential minutes of a call with a competitor on 17.7.2017, Doc ID268.

268               Confirmed non-confidential minutes of a call with a market participant on 12.7.2017, Doc ID323.

269               Document titled […], page 3, Annex 66.1.1.19 to the Form CO.

270               Document titled […], page 5, Annex 66.1.1.11 to the Form CO.

271               Doc ID1658-71506 [internal document].

272               Replies to question 19 of Q13 – Questionnaire to Competitors (Phase II), Doc ID3224.

273               The suppliers usually make recourse to third party logistic providers and suppliers located in similar regions should then have access to similar logistic solutions.

274               M.7155 – SSAB/Rautaruukki, paragraph 67. The assessment in the case concentrated on  those products and not, for instance, on GS. See M.7155 – SSAB/Rautaruukki, paragraph 68.

275               See  M.2382  –  Usinor/Arbed/Aceralia.  See  also  M.7762  –  ArcelorMittal/Financial  Entities/Grupo Condesa, paragraph 30.

276               See replies to Question 11 of Q6 – Questionnaire to Tube Customers, Doc ID3311. See also    replies to Question 12 of Q5 – Questionnaire to Tube Producers, Doc ID2805.

277               M.1014 – British Steel/Europipe.

278               M.1014 – British Steel/Europipe.

279               See,  for  instance,  M.7461  –  AMDS  Italia/CLN/JV,  paragraph  37;  M.7155     –  SSAB/Rautaruukki, paragraph 115; and M.4137 – Mittal/Arcelor, paragraph 80. See also M.2382 – Usinor/Arbed/Aceralia, paragraph 107, where the Commission concluded France constituted a distinct market but left it open, for instance, whether Spain and Portugal constituted distinct markets or belonged to the same market.

280               Official Journal C 31, 5.2.2004, pp. 5-18.

281               Horizontal Merger Guidelines, paragraph 22.

282               ThyssenKrupp  and  Tata  Steel  signed  on  20  September  2017  a  memorandum  of understanding to combine their European steel activities in a 50/50 joint venture.

283               See  inter  alia  Commission  decision  of  2  July  2008  (Case  COMP/M.4942  –       Nokia/Navteq)  at paragraphs 260 and 261; Commission decision of 14 May 2008 (Case COMP/M.4854 – TomTom/Tele Atlas), at paragraphs 187, 188 and Commission decision of 23 November 2011 (Case COMP/M.6203 – WesternDigital Ireland/Viviti technologies) at paragraphs 18–44.

284               See, for instance, the judgment of the Court of First Instance of 6 June 2002 in Case  T-342/99 Airtours plc v Commission [2002] ECR II-2585, paragraph 82 (‘the level of competition obtaining in the relevant market at the time when the transaction is notified is a decisive factor in establishing whether a collective dominant position has been created for the purposes of Regulation No 4064/89’), as well as judgment of the Court of First Instance of 19 May 1994 in Case T-2/93 Air France  v  Commission [1994]  ECR  II-323,  paragraphs  70,  71  and  72;  the  judgment  of  the  Court  of  First  Instance  of  8 July 2003 in Case T-374/00 Verband der  freien  Rohrwerke  and  Others  v  Commission  [2003]  ECR II-2275,   paragraph   170;   the   judgment   of   the    General   Court   (Sixth   Chamber)   of      13 September 2010 in Case T- 279/04 Éditions Odile Jacob v Commission, [2010] ECR 0000, paragraph 327.

285               See point 9 of the Guidelines on the assessment of horizontal mergers under the Council  Regulation on the control of concentrations between undertakings, (OJ C 31, 5.2.2004, p. 5) (the ‘Horizontal Merger Guidelines’).

286               Horizontal Merger Guidelines, paragraph 9.

287               […], Doc ID2431-1330.

288               […]; doc id 001658-068722AM.

289               […].

290               […].

291               The high value and cash-flow generating ability of the Ilva assets seems supported by the fact, that during the Extraordinary administration and before, significant capital investments were made at Ilva to improve/develop the industrial capacity and address requirements of the AIA (Integrated Environmental Authorisation). The capital investment between […], during which period Ilva was managed by the Italian State amounted to EUR […]. Between […] under the Riva management capital expenditure amounted to EUR […]. […].

292               Response to RFI 11 of September 21, 2017 – […], Doc ID1659-37728.

293               It must be  clarified  that the  necessary investments fall in  two  categories,  that is the    remediation of environmental liabilities, for which Ilva has made and is bound to continue to make significant investments, and investments to be made by the acquirer of the assets to progress on meeting environmental standards to the extent that the imposed production cap can be lifted.

294               […], Doc ID001658-068722.

295               […]. Source: […], Doc ID2431-1330.

296               ArcelorMittal’s internal document, […], Doc ID2431-1330, […].

297               See for instance, […]; Reply to RFI 11.

298               HMG, paragraph 14.

299               Notifying Party's Reply to the SO, paragraph 93.

300               See Section 5.1.3.1.

301               Sales of slabs by EEA producers to third parties are relatively rare. And while some HR producers in the EEA import slabs for hot rolling into the EEA, such imports typically consist of such producers own slab production outside the EEA. See recitals (611)-(612).

302               See Section 5.3.

303               See Reply to RFI 18, Q12.

304               See, for example M.6471 – Outokumpu/Inoxum, para 282ff. In that case, which assessed a merger in the flat stainless steel industry, the Commission considered merchant market sales as well as capacity and production shares at the EEA level at each level of the value chain. Moreover, sales shares in cold- rolled stainless steel presented in that case included wholesale captive wholesale sales to integrated steel service centres of EEA producers as well as external sales of cold-rolled products to independent distributors and service centres as well as large customers purchasing directly from steel mills.

305               Consumption shares for  CR  are  computed  following the  same  approach.  For  both HR and  CR the exports of EEA producers are based on actual data for Ilva and ArcelorMittal and estimates for the other EEA suppliers. Exports by competitors are estimated by distributing the residual part of exports (once discounted for the actual volumes of the Parties) across the competitors on the basis of the competitors' production shares (production share computed excluding the Parties). In any case, the market share of the Parties would not change depending on which methodology is used to attribute exports to the EEA competitors.

306               The Notifying Party also argues that vertical effects are unlikely because of re-rollers' ability  to absorb upstream price increases in their margins and because of limited downstream diversion to the merged entity. The Commission acknowledges this and has not raised objections due to vertical effects.

307               See Section 9.4.4.

308               See Section 9.4.5.7.

309               Notifying Party's  submission  of 9  February 2018,  titled  ‘Adjusted  capacity shares are not materially influenced by “double counting”’.

310               Commission's Guidelines on the assessment of horizontal mergers under the Council Regulation  on the control of concentrations between undertakings (the 'Horizontal Merger Guidelines'), OJ C 31,  5.2.2004, p. 5, paragraph 22.

311 Horizontal Merger Guidelines, paragraph 17.

312 Horizontal Merger Guidelines, paragraph 24.

313               Horizontal Merger Guidelines, paragraph 25.

314               Horizontal Merger Guidelines, paragraph 27.

315               Horizontal Merger Guidelines, paragraph 28.

316               Horizontal Merger Guidelines, paragraph 31.

317               Horizontal Merger Guidelines, paragraph 33.

318  Horizontal Merger Guidelines, paragraph 36.

319  Horizontal Merger Guidelines, paragraph 60.

320               Horizontal Merger Guidelines, paragraph 67.

321               Horizontal Merger Guidelines, paragraph 68.

322               Sales shares for 2016 are computed as follows. For AM, Ilva, Marcegaglia, imports and market size: actual data provided by the Parties. For competitors the Parties only provided sales data up to September/October 2016. Sales in the missing months have been then attributed to competitors on the basis of their sales during the period January/September 2016. As the Parties' sales and the market supply are based on actual data the attribution of sales to competitors has no impact on the combined share of the merging Parties.

323               The Notifying Party was given access to the confidential computation in the data room.

324               Data from market reconstruction and the Notifying Party.

325               Consumption shares for EEA are computed as follows. For EEA producers it is the  HR/CR production net of extra EEA exports. For Imports they are the sales from non-EEA producers.

326               Sales share for Southern Europe in 2016 are computed as follows. For AM, Ilva, Marcegaglia and Imports they are based on actual data provided by the Parties. Data on the market size (that is total sales) was available for the full period of 2016, but data on sales of individual EEA competitors were available only up to September/October 2016. Competitors' sales for the remaining months of 2016  have been computed by attributing the part of the market size not already covered by AM, Ilva, Marcegaglia and Imports to the EEA competitors proportionally to their sales shares during the period January-September 2016. It should be noted that as the Parties' sales and the market size are based on actual figures, the attribution of sales to competitors has no impact on the combined share of the Parties. Southern Europe is defined as Italy, Spain, Portugal, Greece and Southern France. Sales in Southern France are those sales made to customer located south of Lyon (for the Parties), or 10% of the sales in France (for the competitors).

327               Consumption  shares  for  Southern  Europe  are  computed  as  follows.  For  Ilva  and  other  Southern producers it is the HR/CR production minus the exports to extra-EEA countries and the sales to Northern Europe. For AM it is the production in its Southern European facilities minus exports to extra-EEA countries and sales to Northern Europe (from the Southern European facilities), plus sales from Northern European plants to Southern Europe. For other Northern European producers, they are the sales from their plants to Southern Europe. Data for Southern France are proxied with 10% of sales to France. For Imports they are the sales of HR/CR from non-EEA producers to the Southern Europe (Southern France again account for 10% of the sales in France). Southern Europe is defined as: Italy, Spain, Greece, Portugal, Southern France (south of Lyon).

328               Metinvest imports slabs from its mother company's facility in Ukraine and hot-rolls them in   its facility located in northern Italy.

329               See, for instance, Form CO, paragraph 834.

330               Form CO, paragraph 835.

331               Confirmed non-confidential minutes of a call with a competitor on 17.7.2017, Doc ID1776.

332               Confirmed non-confidential minutes of a call with a competitor on 18.7.2017, DocID443.

333               Confirmed non-confidential minutes of a call with a competitor on 17.7.2017, Doc ID1776.

334               Confirmed non-confidential minutes of a call with a competitor on 6.4.2017, Doc ID92.

335               Confirmed non-confidential minutes of a call with a competitor on 25.7.2017, Doc ID2938.

336               The classification of the products into specialty and commodity was provided by the Parties.

337               The Notifying Party's reply to Question 27 of RFI 3 of 19 June 2017 (Annex 27.1). The Commission also notes that about […]% of ArcelorMittal's HR sales are classified as ‘No equivalent’ without specifying any EN Grade and claiming that other producers like Ilva would not be able to replicate the same product.

338               […], the Notifying Party's reply to question 26 of RFI 26 of 24 November 2017 (Annex 26).

339               See Horizontal Merger Guidelines, paragraph 28, footnote 32, where geographic location is listed as an example of a parameter of differentiation.

340               Confirmed non-confidential minutes of a call with a customer on 12.6.2017, Doc ID1744.

341               Confirmed non-confidential minutes of a call with a market participant on 12.7.2017, Doc ID323.

342               Confirmed non-confidential minutes of a call with a competitor on 6.4.2017, Doc ID92.

343               Confirmed non-confidential minutes of a call with a customer on 12.6.2017, Doc ID1744.

344               Commission's calculations based on the Notifying Party's data (Annex 7.10 of the Form CO).

345               See, for instance, replies to question 44 of Q3 – Questionnaire to customers (industrial), Doc ID2800, 814, 1347, 1466, 1872, 3555, 3620 and 3621; and to question 44 of Q4 – Questionnaire to customers (SSCs), Doc ID2804 where the majority of industrial customer respondents (Q3) and the majority of SSCs respondents (Q4) indicated AM and Ilva to compete closely in Italy and its surroundings.

346               Reply to question 44.1 of Q4 – Questionnaire to customers (SSCs), Doc ID2804.

347               Reply to question 44.1 of Q3 – Questionnaire to customers (industrial), Doc ID2800.

348               [Parties' submission].

349               See, for example, the Notifying Party's Submission of 4 December 2017 […].

350               Reply to the SO, paragraph 112.

351               Marks added by the Commission.

352               Doc ID1661-79314 [internal document].

353               Reply to the SO, paragraph 112.

354               Doc ID1661-79219 [internal document].

355               The Notifying Party's external consultants were able to examine the Commission's quantitative analysis during the data room procedure.

356               Southern Europe including Italy, Spain, Greece and Portugal. Due to the level of aggregation of the data it was not possible to include Southern France in the model. However, the inclusion of France does not change the Commission's conclusions.

357               Reply to the SO, sections III.B.1 and III.B.2.

358               See evidence made available in the Data Room held from 1 March to 2 March 2018. The Notifying Party argued that […]. While the Commission acknowledges that ArcelorMittal’s data could only be checked against actual customer data for a limited number of customers due to data availability, […].

359               Replies to question 44 of Q4 – Questionnaire to customers (SSCs), Doc ID2804.

360               Replies to question 44 of Q4 – Questionnaire to customers (SSCs), Doc ID2804.

361               Replies to question 44 of Q3 – Questionnaire to customers (industrial), Doc ID2800, 814,   1347, 1466, 1872, 3555, 3620 and 3621.

362               Reply to question 39.1 of Q3 – Questionnaire to customers (industrial), Doc ID2800.

363               Reply to question 39 of Q3 – Questionnaire to customers (industrial), Doc ID2800, 814, 1347, 1466, 1872, 3555, 3620 and 3621.

364               Reply to question 43.1 of Q4 – Questionnaire to customers (SSCs), Doc ID2804.

365               Reply to question 43.1 of Q4 – Questionnaire to customers (SSCs), Doc ID2804.

366               Replies to questions 48-51 of Q1 – Questionnaire to competitors (EEA suppliers of flat   carbon steels), Doc ID2798.

367               Form CO, Annex 1.3, p. 10.

368               Form CO, table 84.

369               See Table 22 and Table 23.

370               Doc ID3294-1251 […].

371               See Section 8.1.2.3.

372               […]. Doc ID1469-88226 [internal document].

373               Doc ID55-3 [internal document].

374               Doc ID1661-79313 [internal document], p. 15.

375               Doc ID1469-7993 [internal document].

376               Doc ID1661-17513 […].

377               Doc ID617-3020 […].

378               Doc ID617-149 […].

379               Doc ID555-1213 […].

380               Doc ID1661-6299 [internal document].

381               Doc ID1469-1344 [internal document].

382               Recitals (576) to (579).

383               Doc ID1469-36751 [internal document].

384               The Notifying Party's reply to RFI 7, Annex 66.1 – [internal document].

385               See recital (1289)(2).

386               See recital (1286).

387               Doc ID1469-52600 [internal document].

388               Doc ID1660-17812 […].

389               […] (Doc ID1469-88649, [internal document]); […] (Doc ID1659-91761, [internal document]).

390               Doc ID163-12 […].

391               Replies to question 63 of Q1 – Questionnaire to competitors (EEA suppliers of flat carbon steel), Doc ID2798 and  confirmed  non-confidential  minutes  of  a  call  with  a  competitor  on  6.4.2017,  Doc ID92.

392               Replies to Question 50 of Q4 – Questionnaire to customers (SSCs), Doc ID2804.

393               Courtesy translation from Italian: 'pensa che Ilva abbia una politica aggressiva di prezzo per  i prodotti sopra indicati quando vende in Italia'.

394               Replies to Question 50 of Q3 – Questionnaire to customers (industrial), Doc ID2800, 814,  1347, 1466, 1872, 3555, 3620 and 362; and replies to question 50 of Q4 – Questionnaire to customers (SSCs),     Doc ID2804.

395               Replies to Questions 49 and 50 of Q3 – Questionnaire to customers (industrial), Doc ID2800.

396               Courtesy translation from Italian: 'Da metà degli anno '90 Ilva ha avuto una funzione calmierante sul mercato italiano a vantaggio degli utilizzatori finali e delle loro esportazioni. Pertanto l'acquisizione di Ilva da parte di Arcelor induce a pensare ad un futuro aumento di prezzi in Italia… '. Replies to question 84 to Questionnaire 4 to SSCs, Doc ID2804.

397               See Table 2.

398               Reply to the SO, paragraph 153.

399               See recitals (107) - (110) and Table 2.

400               Annex 2.1 […] of the Notifying Party's response to the Article 6(1)(c) decision.

401               The Notifying Party's reply to question 2 of RFI 28.

402               The Notifying Party's reply to RFI 25, Doc ID2431-1330 ([…]).

403               Doc ID555-67 […].

404               Doc ID1658-78487 [internal document].

405               Doc ID1659-48985 [internal document].

406               Doc ID1469-11066 [internal document].

407               Doc ID1469-66627 [internal document].

408               Doc ID1660-37407 [internal document].

409               Doc ID2339, RFI 20, Annex 1.5.

410               The Notifying Party's reply to RFI 7, Annex 43-3, […].

411               Doc ID1658-92216 [internal document].

412               Doc ID1659-74014 [internal document].

413               Reply to question 53.1 of Q3 – Questionnaire to customers (industrial), Doc ID2800.

414               This would happen directly via either the base price + extra formula or the price indexation for long term contracts, or indirectly as competitive condition in the commodity markets might be reflected in  the price negotiations of more added value products (for example the specialised press usually discusses the price developments and competitive conditions in the commodity product space).

415               Confirmed non-confidential minutes of a call with a market participant on 25.7.2017, Doc ID2062.

416               Reply to the SO, Section E.1.b and Annex 3.D.

417               Reply to the SO, paragraph 223 and subsequent.

418               […]; Doc ID1660-54969 [internal document].

419               […]; Doc ID1659-75218 [internal document].

420               Doc ID1539-1 […].

421               […]; Doc ID1660-56554 [internal document].

422               The Notifying Party further contested that the Commission does not include in its computation of HR spare capacity smaller suppliers. In this respect, the Commission notes that the inclusion of smaller suppliers would not change the Commission's conclusion. Furthermore, ArcelorMittal data shows a  very low utilisation rate for smaller suppliers. As the Commission found a sizeable discrepancy in the utilisation rates between the figures provided by ArcelorMittal for the main EEA competitors, and the figures provided by the main EEA competitors in the market reconstruction, the Commission considers it likely that such discrepancy could also be present for smaller suppliers, or that they could be affected by other bottlenecks,

423               See Figure 75.

424               Horizontal Merger Guidelines, paragraph 24.

425               See Notifying Party's Reply to question 1 of RFI 26.

426               Commission's computation based on data provided by the Notifying Party.

427               Reply to the SO, Section E.2.

428               It is essentially the  difference between Scenario  1  and  Scenario  2  in the capacity share computation described in recitals (490)-(492).

429               Notifying Party's Reply to the SO, Figure 57.

430               The Notifying Party's Reply to the Article 6(1)(c) decision, page 62, Response to SO, paragraph 240 et seq.

431               The Notifying Party's Reply to the Article 6(1)(c) decision, page 63.

432               The Notifying Party's Reply to the Article 6(1)(c) decision, page 65.

433               The Notifying Party’s Reply to the Letter of Facts, page 40.

434               Reply to the SO, paragraph 250.

435               The Notifying Party's Reply to the Article 6(1)(c) decision, page 67.

436               Reply to the SO, paragraph 247.

437               The Notifying Party's Reply to the Article 6(1)(c) decision, page 76. […].

438               Reply to the SO, paragraph 256 et seq.

439               The Notifying Party’s Reply to the Letter of Facts, page 34.

440               Reply to the SO, paragraph 259 et seq.

441               The Notifying Party’s Reply to the Letter of Facts, page 36.

442               Steel  Business  Briefing  ‘European buyers require  established  imports:  Macrometal’,  12  May 2017 Doc ID1469-31427 […].

443               Reply to the SO, paragraph 241 and Eurostat's trade statistics.

444               Reply to the SO, paragraph 250.

445               Doc ID3608-3609.

446               Contrary to the Notifying Party's argument (Reply to the SO, paragraph 245), the high level   of relative domestic HR prices during 2006-2008 cannot be solely explained by an increase in the shipping costs of steel and raw materials to the EEA. While the Baltic Dry Index used by the Notifying Party displays a steep increase starting in 2006 until the end of 2008, a period during which the index increased by about five times, the relative prices shown in Figure 77 display only an increase during 2008 relative to the previous years.

447               Commission Implementing Regulation (EU) 2017/1795 of 5 October 2017 imposing a    definitive anti- dumping duty on imports of certain hot-rolled flat products of iron, non-alloy or other alloy steel originating in Brazil, Iran, Russia and Ukraine and terminating the investigation on imports of certain hot-rolled flat products of iron, non-alloy or other alloy steel originating in Serbia, L 258, 6.10.2017, p. 24, recital 423.

448               Form CO, paragraphs 568 – 581 and the Notifying Party's reply to question 11 of RFI 23 dated 17 November 2017.

449               Form CO, paragraph 568.

450               The Notifying Party's reply to question 11 of RFI 23 dated 17 November 2017.

451               Form CO, paragraph 581.

452               The Notifying Party's reply to question 11 of RFI 23 dated 17 November 2017.

453               The Notifying Party's reply to question 11 of RFI 23 dated 17 November 2017.

454               Form CO, paragraph 576.

455               Form CO, paragraph 577.

456               Form CO, paragraph 580.

457               The Notifying Party's reply to question 11 of RFI 23 dated 17 November 2017.

458               Reply to the SO, paragraph 271.

459               Reply to question 37 of Q11 - Questionnaire to Industrial Customers (Phase II), Doc ID3225.

460               The Notifying Party's Reply to the SO, paragraph 270 and following.

461               Doc ID4255.

462               Confirmed non-confidential minutes of a call with a competitor on 25.7.2017, Doc ID2938.

463               Confirmed non-confidential minutes of a conference call with an industrial customer on 12.6.2017, Doc ID253.

464               Confirmed non-confidential minutes of a call with a non-EEA producer on 1.9.2017, Doc ID3128.

465               Confirmed non-confidential minutes of a call with a non-EEA producer on 1.9.2017, Doc ID3128.

466               Reply to question 19 of Q2 - Questionnaire to Importers, Doc ID2799.

467               Confirmed non-confidential minutes of a call with a market participant on 6.12.2017, Doc ID3686.

468               Confirmed non-confidential minutes of a call with a customer on 31.8.2017, Doc ID2145.

470               Metal Bulletin 12/02/2018 Indian domestic HRC prices stable despite supply constraints, Doc ID4251.

471               Confirmed non-confidential minutes of a call with a non-EEA producer on 6.12.2017, Doc ID3484.

472               SBB briefing 16/02/2018 Asia hot-rolled coil prices remain stable on holiday lull, Doc ID4257.

473               Imports  data  from  Eurostat  are  provided  aggregated  at  the  extra  EU28  level.  For  this reason the considered flow of imports is from outside the EU28 to the EU28 countries. The same analysis conducted at the extra EEA level would not materially change the results.

474               See Section 9.4.5.3.

475               Contrary to the Notifying Party's argument (Reply to the SO, paragraph 278), the    observed cyclicality of imported HR volume in the EEA is not in contradiction with the relation with relative prices  observed in Figure 77. Rather, they may both be related to other factors that cause the cyclical pattern.

476               Confirmed non-confidential minutes of a conference call with a customer on 12.7.2017, Doc ID253.

477               Reply to the Letter of Facts, paragraph 92.

478               Reply to the SO, paragraph 250 and paragraph 274 and following.

479               See recital (665).

480               A similar reasoning would apply taking into account the price in China rather than the raw material costs, in line with the argument of the Notifying Party. Figure 77 shows that that the relative price between Southern European HR price and Chinese HR price significantly vary over time, indicating  that also the relative spreads will be subject to an even higher variation over time (as raw material costs are the same for both price series).

481               This analysis is based predominantly on data that pre-dates the recent anti-dumping measures.

482               The Notifying Party's criticisms in relation to the    Commission's quantitative analysis are addressed in the Annex 1 to this Decision.

483               The Notifying Party's Reply to the Letter of Facts, paragraph 99 and Table 5.

484               SBB DAILY BRIEFING 10/01/2018 (Doc ID3636).

485               SBB DAILY BRIEFING 8/01/2018 (Doc ID3635).

486               SBB HRC imports into EU fall to near three-year low Monday, 18 December 2017 (Doc ID3643).

487               SBB HRC imports into EU fall to near three-year low Monday, 18 December 2017 (Doc ID3643).

488               SBB  Indian  cold-rolled  continued  to  dominate  EU  market  in  Oct  Thursday,  21   December  2017 (Doc ID3649).

489               SBB  Importers  await  direction  from  European  coil  producers  Wednesday,  01      November  2017 (Doc ID3638).

490               SBB Southern European coil mills contemplating increases Monday, 03 July 2017 (Doc ID3639).

491               See, for example, ArcelorMittal's Response to the Article 6(1)(c) decision dated 17 November 2017, paragraph 149 et seq.

492               S&P Global Platts, Global Market Outlook, December 2017, page 14.

493               The green and red marks on the slide were introduced by the Commission.

494               Commission Implementing Regulation (EU) 2016/1778 of 6 October 2016 imposing a provisional anti- dumping duty on imports of certain hot-rolled flat products of iron, non-alloy or other alloy steel originating in the People's Republic of China (OJ L 272, 7.10.2016, p. 33).

495               Registration  requirement  imposes  to  customs  to  register  the  imports  from a  given  country. In the context of an anti-dumping procedure, registration implies that subsequent tariffs the procedure may lead to are imposed retroactively for the registered imported volumes.

496               The comparison of Figure 88 with Figure 87 should consider the effect of a time lag. Figure   87 reports the prices of landed imports into Europe or Italy, the price of these landed imports had been negotiated at least three/four months before. Figure 87 reports the prices as monitored by the data provider SBB and these reflects the prevailing price at the moment of the negotiation not of delivery.

497               Doc   ID3735.   ‘ArcelorMittal   increase   European   coil   offer   prices’,   Platts       Breaking   News, 12 December 2017.

498               Doc  ID3637,  ‘Italian  mills  hike  coil  prices  as  expected,  sources  say’,  SBB  Daily  Briefing, 15 January 2018.

499               The Notifying Party's reply to RFI 7, Annex 66.1, document '20170503_Europe Flat Products    - Q1'17 BAR Main final', page 9.

500               Doc ID1469-6401 […].

501               Doc ID1469-98344 […].

502               Doc ID1661-50763 […].

503               Page 3 of document titled […], Doc ID158-10.

504               Replies  to  questions  20  and  21  of  Q12  -  Questionnaire  to  Steel  Distributors  /  SSCs  (Phase  II), Doc ID3223.

505               Replies to question 25.2 of Q11 - Questionnaire to Industrial Customers (Phase II), Doc   ID3225, 3624 and 3625.

506               Confirmed non-confidential minutes of a call with a competitor on 6.4.2017, Doc ID92.

507               Confirmed non-confidential minutes of a call with a customer on 28.11.2017, Doc ID3780.

508               Replies to  question 61  of Q3 –  Questionnaire  to  customers (industrial)  (Phase  I), Doc ID2800, 814, 1347, 1466, 1872, 3555, 3620 and 3621 and replies to question 61 of Q4 – Questionnaire to customers (SSCs) (Phase I), Doc ID2804.

509               Replies to question 61.1 of Q3 – Questionnaire to customers (industrial) (Phase I), Doc ID2800, 814, 1347, 1466, 1872, 3555, 3620 and 3621 and replies to question 61.1 of Q4 – Questionnaire to customers (SSCs) (Phase I), Doc ID2804.

510               Reply to question 69.1 of Q4 – Questionnaire to customers (SSCs) (Phase I), Doc ID2804.

511               Replies to  question 69  of Q3 –  Questionnaire  to  customers (industrial)  (Phase  I), Doc ID2800, 814, 1347, 1466, 1872, 3555, 3620 and 3621 and question 69 of Q4 – Questionnaire to customers (SSCs) (Phase I), Doc ID2804.

512               Replies to question 69.1 of Q3 - Questionnaire to customers (industrial) (Phase I), Doc ID2800.

513               Replies to question 68.1 of Q3 – Questionnaire to customers (industrial) (Phase I), Doc ID2800.

514               Replies to questions 23 and 24 of Q11 - Questionnaire to Industrial Customers (Phase II), Doc ID3225.

515               Replies to question 23 of Q11 - Questionnaire to Industrial Customers (Phase II), Doc ID3225.

516               Confirmed non-confidential minutes of a call with a trader on 5.12.2017, Doc ID3517.

517               Replies to question 69.1 of Q4 – Questionnaire to Customers (SSCs) (Phase I), Doc ID2804.

518               Replies to question 69.1 of Q3 – Questionnaire to customers (industrial) (Phase I), Doc ID2800.

519               Replies to question 23 of Q11 - Questionnaire to Industrial Customers (Phase II), Doc ID3225.

520               Replies to question 69.1 of Q3 – Questionnaire to customers (inudstiral) (Phase I), Doc ID2800.

521               Replies to question 68.1 of Q3 – Questionnaire to customers (inudstiral) (Phase I), Doc ID2800.

522               Confirmed minutes of a call with a market participant on 12.7.2017, Doc ID323.

523               Confirmed non-confidential minutes of a call with a market participant on 6.12.2017, Doc ID3686.

524               Commission Implementing Regulation (EU) 2017/1795 of 5 October 2017, recital 463.

525               SBB daily briefing 31/01/2018 S.EU HRC prices move up aggressively in tight market, Doc ID4254.

526               The Notifying Party’s Reply to the Letter of Facts, page 42.

527               Replies to question 63 of Q4 – Questionnaire to customers (SSCs) (Phase I), Doc ID2804 and replies to question 63 of Q3 – Questionnaire to customers (industrial) (Phase I), Doc ID2800, 814, 1347, 1466, 1872, 3555, 3620 and 3621

528               Replies to question 84 of Q1 – Questionnaire to competitors (EEA suppliers of flat carbon steel), Doc ID2798.

529               Replies to question 66 of Q3 – Questionnaire to customers (industrial), Doc ID2800, 814,   1347, 1466, 1872, 3555, 3620 and 3621; and replies to question 66 of Q4 – Questionnaire to customers (SSCs),    Doc ID2804.

530               Replies to question 16 of Q12 – Questionnaire to Steel Distributors / SSCs (Phase II), Doc ID3223.

531               Replies to question 19 of Q11 – Questionnaire to Industrial Customers (Phase II), Doc ID3225.

532               See recital (722) et seq.

533               [Commercial information].

534               Contrary to the Notifying Party's argument (Reply to the SO, paragraphs 279-284), the fact that, even during a period characterised by very low import prices, the vast majority of customers purchased only  a limited portion of their demand from imports suggests that the ability of customers to switch or increase their purchases from imports may be limited in the post-Transaction scenario.

535               Doc ID1661-079219.

536               Doc ID2339, RFI 20, Annex 1.5.

537               The Notifying Party’s reply to the Letter of Facts, page 39.

538               […], Doc ID1658.

539               Doc ID2339, RFI 20, Annex 1.5.

540               Reply to the SO, paragraphs 281-282.

541               Reply to the Letter of Facts, paragraph 95.

542               Replies to question 12.1 of Q2 - Questionnaire to Importers, Doc ID2799.

543               Confirmed non-confidential minutes of a call with a trader on 5.12.2017, Doc ID3517.

544               Minutes  of  a  call  between  the  Commission's  case  team  and  an  anonymous  market participant on 15 December 2017, Doc ID3574.

545               Replies to question 33.1 of Q3 – Questionnaire to customers (industrial) (Phase I), Doc ID2800, 814, 1347, 1466, 1872, 3555, 3620 and 3621.

546               Confirmed non-confidential minutes of a call with a customer on 31.8.2017, Doc ID2145.

547               Replies to question 19 of Q12 – Questionnaire to Steel Distributors / SSCs (Phase II), Doc  ID3223 and question 22 of Q11 - Questionnaire to Industrial Customers (Phase II), Doc ID3225, 3624 and 3625.

548               See, for example, replies to question 19 and 19.1 of Q12 – Questionnaire to Steel Distributors / SSCs (Phase II), Doc ID3223 and replies to question 22.1 of Q11 - Questionnaire to Industrial Customers (Phase II), Doc ID3225, 3624 and 3625.

549               Replies to question 20.5 of Q11 - Questionnaire to Industrial Customers (Phase II), Doc ID3225.

550               Replies to question 14.1 of Q12 – Questionnaire to Steel Distributors (SSCs) (Phase II), Doc ID3223.

551               Replies to question 63.1 of Q4 - Questionnaire to customers (SSCs) (Phase I), Doc ID2804.

552               See, for example, replies to question 22 of Q11 - Questionnaire to Industrial Customers (Phase II), Doc ID3225, 3625 and 3625; and question 62 of Q3 - Questionnaire to Industrial Customers (Phase I), Doc ID2800, 814, 1347, 1466, 1872, 3555, 3620 and 3621.

553               Confirmed non-confidential minutes of a conference call with an industrial customer on 25.4.2017, Doc ID102.

554               Confirmed  non-confidential  minutes  of  a  conference  call  with  a  market  participant  on 3.10.2017 Doc ID2158.

555               Confirmed non-confidential minutes of a conference call with a distributor on 28.4.2017, Doc ID272.

556               In this respect, a customer explained: 'Besides the long lead time for import from outside the EEA, another possible big issue is the availability of credit lines. While substantially increasing the import,  so the need to pay by L/C [letter of credit], this might become an issue.' Reply to question 33.1, 33.2,

33.3 of Q3, Doc ID2800.

557               Some respondents have also indicated that even if the quality of imports might be  acceptable, handling of any quality issues with importers is difficult. See replies to question 33, 62, 64 and 65 of Q3 – Questionnaire to customers (industrial); and replies to questions 33, 62, 64 and 65 of Q4 –  Questionnaire to customers (SSCs).

558               Replies to question 17.5 of Q12 – Questionnaire to Steel Distributors / SSCs (Phase II), Doc ID3223.

559               Replies to question 20.5 of Q11 - Questionnaire to Industrial Customers (Phase II), Doc ID3225.

560               Replies to question 20.5 of Q11 - Questionnaire to Industrial Customers (Phase II), Doc ID3225.

561               Reply to Letter of Facts, paragraph 94.

562               See recitals (581)-(584).

563               The  Notifying  Party's  reply  to  question  3  of  RFI  26  as  well  as  replies  to  question  13  of Q13 - Questionnaire to Competitors (Phase II), Doc ID3224.

564               Doc ID1469-61892 […].

565               […]. Doc ID2339, RFI 20, Annex 1.5.

566               [Parties' submissions].

567               See recital (430).

568               See Section 11.1.

569               Doc ID1658-62811 […].

570               Doc ID1469-88103 […].

571               Doc ID1658-92377 […].

572               Doc ID1658-19574 […].

573               Doc ID1661-30943 […].

574               Reply to the SO, paragraphs 303-310.

575               See Section 8.1.2.5.

576               Response to the Article 6(1)(c) decision dated 17 November 2017, p. 5-6.

577               Form CO, p. 232.

578               Sales share for 2016 are computed as follows. For AM, Ilva, Marcegaglia, imports and market size: actual data provided by the Parties. For competitors the Parties only provided sales data up to September/October 2016. Sales in the missing months have been then attributed to competitors on the basis of their sales during the period January/September 2016. As the Parties sales and the market supply are based on actual data the attribution of sales to competitors has no impact on the combined share of the merging parties.

579               See Section 9.3.1 which describes how Scenario 1 and Scenario 2 have been calculated.

580               See Section 7.2.5: Conditions of competition are at least differentiated within the EEA.

581               Sales share for 2016 are computed as follows. For AM, Ilva, Marcegaglia, imports and market size: actual data provided by the Parties. For competitors the Parties only provided sales data up to September/October 2016. Sales in the missing months have been then attributed to competitors on the basis of their sales during the period January/September 2016. As the Parties sales and the market supply are based on actual data the attribution of sales to competitors has no impact on the combined share of the merging Parties.

582               [Parties' submissions].

583               See, for instance, replies to question 40 of Q3 – Questionnaire to customers (industrial), Doc ID2800 and to question 44 of Q4 – Questionnaire to customers (SSCs), Doc ID2800 where the majority of industrial customer respondents (Q3) and the majority of SSCs respondents (Q4) indicated AM and Ilva to compete closely in Italy and its surroundings.

584               Replies to question 44.1 of Q4 – Questionnaire to customers (SSCs), Doc ID2804.

585               Replies to question 95 of Q11 – Questionnaire to Industrial Customers (Phase II), Doc ID3225.

586               Arvedi, produces CR at its Cremona plant where it produces both commodity and higher value CR products. According to the Notifying Party, it has a CR capacity of c.[…], of which only […]% is effectively used. According to information submitted by the Notifying Party, Arvedi was also due to start operating a new CR facility in Trieste in 2017 with a capacity of […].

587               See  also  confirmed  non-confidential  minutes  of  a  conference  call  with  a  customer on 12.6.2017; Doc ID1744.

588               […]. Doc ID1469-88226 […].

589               See replies to Q3 – Questionnaire to Customers (Industrial), Doc ID2800 and Q1 – Questionnaire to Competitors, Doc ID2798.

590               Plans to increase CR capacity are either already taken into account in the   Commission's computations, or they are not located in the area where the anti-competitive effects of the Transaction will primarily materialise.

591               […] [confidential].

592               […] [confidential].

593               Response to the Article 6(1)(c) decision, dated 17 November 2017, p. 6.

594               […] [confidential].

595               Replies to question 88 of Q11 – Questionnaire to Industrial Customers (Phase II), Doc ID3225.

596               Replies to  question 85  of Q11- Questionnaire  to  Industrial Customers (Phase  II), Doc   ID3225. The other 58% of respondents indicated that they were not importing/found the question to be inapplicable to their circumstances.

597               Measures in force against Russia and the PRC. See, inter    alia, Commission Implementing Regulation (EU) 2015/2325 of 11 December 2015 making imports of certain cold rolled flat steel products originating in the People's Republic of China and the Russian Federation subject to registration.

598               Replies to question 90 of Q11 – Questionnaire to Industrial Customers (Phase II), Doc ID3225.

599               Replies to question 96 of Q11 – Questionnaire to Industrial Customers (Phase II), Doc ID3225.

600               Replies to question 89 of the Q11 – Questionnaire to Industrial Customers (Phase II), Doc ID3225.

601               Replies to question 82 of the Q11 – Questionnaire to Industrial Customers (Phase II), Doc ID3225.

602               Replies to question 85 of Q11 – Questionnaire to Industrial Customers (Phase II), Doc ID3225.

603               Replies to question 85 of Q11 – Questionnaire to Industrial Customers (Phase II), Doc ID3225.

604               Replies to question 92 of Q3 – Questionnaire to customers (industrial), Doc ID2800.

605               Confirmed  non-confidential  minutes  of  a  call  with  a  non-EEA  producer  on  6     December  2017, Doc ID3484.

606               See also confirmed  non-confidential  minutes of a conference call  with a customer on 12    June 2017, Doc ID1744.

607               Replies to question 106 of Q11 – Questionnaire to Industrial Customers (Phase II), Doc ID3225. See also replies to question 79 of Q3 – Questionnaire to customer (industrial), Doc ID2800 and replies to question 85 of Q4 – Questionnaire to customers (SSCs).

608               Replies to question 107 of Q11 – Questionnaire to Industrial Customers (Phase II), Doc ID3225.

609               Replies to question 108 of Q11 – Questionnaire to Industrial Customers (Phase II), Doc ID3225).

610               Replies to question 107 of Q11 – Questionnaire to Industrial Customers (Phase II), Doc ID3225. See also replies to question 85.1 of Q4 – Questionnaire to customers (SSCs).

611               Replies to question 106 of Q1 – Questionnaire to competitors (EEA suppliers of flat carbon steels).

612               Replies to question 79.1 of Q3 – Questionnaire to customers (industrial), Doc ID2800.

613               Replies to question 125 of Q11 – Questionnaire to industrial customers (Phase II), Doc ID2800.

614               See,  for  example,  confirmed  non-confidential  minutes  of  a  conference  call  with  a    customer  on 25 July 2017, Doc ID2062.

615               Replies to question 85.1 of Q4 – Questionnaire to customers (SSCs).

616               See  for  example  confirmed  non-confidential  minutes  of  a  conference  call  with  a      customer  on 4 August 2017, Doc ID2045.

617               Based on information in the Form CO, Table 108.

618               Sales share for 2016 are computed as follows. For AM, Ilva, Marcegaglia, imports and market size: actual data provided by the Parties. For competitors the Parties only provided sales data up to September/October 2016. Sales in the missing months have been then attributed to competitors on the basis of their sales during the period January/September 2016. As the Parties sales and the market supply are based on actual data the attribution of sales to competitors has no impact on the combined share of the merging parties. Southern Europe is defined as Italy, Spain, Portugal, Greece and Southern France. Sales in Southern France are those sales made to customer located south of Lyon (for the Parties), or 10% of the sales in France (for the competitors).

619               Scenario  1:  immediately  available  capacity;  Scenario   2:  available  in  short/medium       term.  See recitals (491) and (492).

620               Data from market reconstruction and Notifying Party.

621               Replies  to  question  61  of  Q11-  Questionnaire  to  Industrial  Customers  (Phase  II),  and  replies  to question 53 of Q12 – Questionnaire to Steel distributors / SSCs (Phase II), Doc ID3223 and 3225.

622               Reply to question 61 of Q11 – Questionnaire to Industrial Customers (Phase II), Doc ID3225.

623               Confirmed non-confidential minutes with a steel supplier, 21 November 2017, Doc ID3183.

624               See, for instance, replies to question 60 of Q11- Questionnaire to Industrial Customers (Phase II) and replies to question 52 of Q12 – Questionnaire to Steel distributors / SSCs (Phase II), Doc ID3223       and 3225.

625               Form CO, paragraphs 834–6.

626               Form CO, paragraph 673.

627               Doc ID1659-73229 […].

628               Doc ID1659-73229 […].

629               Doc ID2454-13 […].

630               Replies to question 50 of Q12 – Questionnaire to Steel distributors / SSCs (Phase II), Doc  ID3223, and replies to question 58 of Q11- Questionnaire to Industrial Customers (Phase II), Doc ID3225. See also replies to questions 49  and 51  of  Q12  –  Questionnaire  to  Steel  distributors  /  SSCs  (Phase  II),  Doc ID3223, and replies to questions 57 and 59 of Q11 – Questionnaire to  Industrial  Customers  (Phase II), Doc ID3225.

631               Replies  to  questions  49  and  51  of  Q12  –  Questionnaire  to  Steel  distributors  /  SSCs  (Phase  II), Doc ID3223, and replies to questions 57 and 59 of Q11 – Questionnaire to  Industrial  Customers  (Phase II), Doc ID3225.

632               The original in Italian reads: ‘La qualitá degli HDG prodotti dallo stabilimento di Novi è  molto buona; ultimamente è un po' peggiorata perché è peggiorata la qualità del materiale HRC - - Gli HDG  prodotti dagli altri stabilimenti Ilva sono meno ‘belli’. See reply to question 49 of Q12 – Questionnaire to Steel distributors / SSCs (Phase II). Doc ID3223.

633               Replies to questions 57–9 of Q11 Questionnaire to Industrial Customers (Phase II). Doc ID3225.

634               Doc ID1469-1976 […].

635               The customer overlap analysis at the location level identifies a common customer only if both  Ilva and ArcelorMittal served that same customer location in 2016. The same analysis at the group  level identifies common customers also when Ilva and ArcelorMittal do not serve the same location but serve different locations of the same customer. The difference between customer location and customer group only applies to customers that purchase steel products at different locations.

636               […].

637               […], Annex 3.2 to Response to the Article 6(1)(c) decision. See also […].

638               […], Annex 3.2 to Response to the Article 6(1)(c) decision.

639               Source: […].

640               Source: […].

641               For  instance,  see  replies to  question 55  of Q11  –  Questionnaire  to  Industrial Customers (Phase II), Doc ID3225; and replies to question 47 of Q12 – Questionnaire to steel distributors / SSCs (Phase II), Doc ID3223. See also replies to questions 41 and 44 of Q3 – Questionnaire to customers (industrial), Doc ID2800, 814, 1347, 1466, 1872 and 1891; and replies to questions 41 and 44 of Q4 – Questionnaire to customers (SSCs), Doc ID2804, where particularly competition in price of GS products but also with regard to other parameters was suggested by customers that took a position.

642               Replies to question 50 of Q1 – Questionnaire to Competitors (EEA suppliers of flat carbon steel), Doc ID2798. While the question concerned GS, the Commission considers it applies to HDG as well, given  that  HDG  constitutes  the  vast  majority of  all  GS  produced  and  sold  in  the  EEA.  See also

confirmed minutes of a call with a competitor on 21 November 2017, where a competitor producing commodity HDG in Southern Europe indicates that both Ilva and ArcelorMittal are its important competitors. Doc ID3183.

643               Form CO, Tables 102 and 131, and reply to question 18 of RFI 19.

644               For 2016, the Notifying Party has suggested the effective capacity of the production lines in  Genova to have been […] with the effective production capacity of […] having been estimated at […] less than for 2017. Production on the line started in 2016.

645               […].

646               Form CO, Annex 7.12.

647               Doc ID1660-98070 […].

648               Doc ID002454-22 […].

649               Table 103 of the Form CO.

650               The Notifying Parties and market reconstruction.

651               See, for intance, […].

652               Doc ID1659-18404 […].

653               Doc ID617-3 (Annex 66.1 to the Form CO, […]).

654               Doc ID617-3 (Annex 66.1 to the Form CO, […]).

655               Such reactions are in line with the reaction of competitors described in paragraph 24 of   the Horizontal Merger Guidelines. On the basis of economic theory, in the context of differentiated markets (including geographic differentiation) or markets with homogeneous products in which excess capacity is not extremely large, the most profitable response by non-merging firms is generally to at least partially follow a post-merger price increase rather than fully defeat it.

656               Annex 3.8.1 to the Form CO.

657               See, for instance, paragraph 31 of the Form CO.

658               See, for instance, replies to question 103 of Q1 – Questionnaire to competitors (EEA suppliers of flat carbon steel), Doc ID2798; replies to question 76 of Q3 – Questionnaire  to  customers  (industrial),  Doc ID2800, 814, 1347, 1466, 1872 and 1891; and replies to question 76 of Q4 – Questionnaire to customers (SSCs), Doc ID2804.

659               Reply to question 76 of Q3 – Questionnaire to customers (industrial). Doc ID2800.

660               Commission  Implementing  Regulation  (EU)  2017/1238  of  7  July  2017  making  imports of certain corrosion resistant steels originating in the People's Republic of China subject to registration (OJ L 177, 8.7.201, p. 39).

661               Commission Implementing Regulation (EU) 2017/1444 of 9 August 2017 imposing a  provisional anti- dumping duty on imports of certain corrosion resistant steels originating in the People's Republic of China (OJ L 207, 10.8.2017, p. 1).

662               See DG TRADE's website:

http://trade.ec.europa.eu/tdi/case details.cfm?ref=com&id=2255&sta=1&en=20&page=1&number=&p rod=steel&code=&scountry=all&ringbjaproceed=Anti-dumping&status=all&measures=all&measure type=all&search=ok&c order=name&c order dir=Up#f ootnote-5.

663               Replies to question 54 of Q11 – Questionnaire to Industrial Customers (Phase II), Doc ID3225; and replies to question 46 of Q12 – Questionnaire to Steel distributors / SSCs (Phase II), Doc ID3223.

664               Replies to question 46 of Q12 – Questionnaire to Steel distributors / SSCs (Phase II), Doc ID3223.

665               Replies to question 45 of Q11 – Questionnaire to Industrial Customers (Phase II), Doc ID3225; and replies to question 53 of Q12 – Questionnaire to Steel distributors / SSCs (Phase II), Doc ID3223.

666               Replies to question 67 of Q3 – Questionnaire to customers (industrial); and replies to question 67 of Q4 – Questionnaire to customers (SSCs). See also replies to questions 62–6 of Q3 – Questionnaire to customers (industrial) and replies to questions  62–6  of  Q4  –  Questionnaire  to  customers  (SSCs). Doc ID2800, 2804, 814, 1466, 1872 and 1891.

667               The original in German reads: ‘Feuerverzinkt ist bereits jetzt schon scwer zu deschaffen, auch im Import.

668               The  original  in Italian reads:  ‘Da  metà  degli  anno  '90 Ilva  ha avuto una  funzione  calmierante sul mercato italiano a vantaggio degli utilizzatori finali e delle loro esportazioni. Pertanto l'acquisizione di Ilva da parte di Arcelor induce a pensare ad un futuro aumento di prezzi in Italia.’

669               Replies to question 87 of Q4 – Questionnaire to customers (SSCs). Doc ID2804.

670               Replies to question 81 of Q3 – Questionnaire to customers (industrial). Doc ID2800, 814,   1347, 1466, 1872 and 1891.

671               Replies to question 62 of Q11 – Questionnaire to Industrial Customers (Phase II), Doc ID3225; and replies to question 54 of Q12 – Questionnaire to Steel distributors / SSCs (Phase II), Doc ID3223.

672               Italy, Spain, Portugal, Greece and 10% of supplies in France.

673               Sales share for 2016 are computed as follows. For AM, Ilva, Marcegaglia, imports and market size: actual data provided by the Parties. For competitors the Parties only provided sales data up to September/October 2016. Sales in the missing months have been then attributed to competitors on the basis of their sales during the period January/September 2016. As the Parties sales and the market supply are based on actual data the attribution of sales to competitors has no impact on the combined share of the merging parties.

674               Data from market reconstruction and Notifying Party.

675               Capacity readily available, see recital (491).

676               Capacity available with limited constraints, see recital (492).

677               Replies to question 56 of Q11 – Questionnaire to Industrial Customers (Phase II), Doc ID3225; and replies to question 48 of Q12 – Questionnaire to Steel distributors / SSCs (Phase II), Doc ID3223. See also replies to question 41 Q3 – Questionnaire to customers (industrial), Doc ID2800, 814, 1466, 1872 and 1891; and replies to question 41 of Q4 – Questionnaire to customers (SSCs), where particularly competition in price of GS products but also with regard to other parameters was suggested by customers that took a position, Doc ID2804.

678               Replies to question 44 of Q4 – Questionnaire to customers (SSCs), Doc ID2804.

679               Replies to question 44 of Q3 – Questionnaire to customers (industrial). Doc ID2800, 814,   1347, 1466, 1872 and 1891.

680               Replies  to  question 53  of Q11  Questionnaire  to  Industrial  Customers  (Phase  II), Doc ID3225; and replies to question 46 of Q12 – Questionnaire to Steel distributors / SSCs (Phase II), Doc ID3223.

681               The original in Italian reads: ‘In una situazione di quasi monopolio, il prezzo verrà influenzato dalle decisioni del nuovo soggetto’.

682               Replies to question 80 of Q3 – Questionnaire to customers (industrial) and replies to question 86 of  Q4 – Questionnaire to customers (SSCs). Doc ID2800, 2804, 814, 1466, 1872 and 1891.

683               The Commission notes that even if a distinction between metallic steel for packaging used for beverage and non-beverage applications were to be made, the Transaction would not result in a significant impediment to effective competition because (i) Ilva is not active in the production of metallic coated steel for packaging used in beverage applications and thus no affected markets arise; and (ii) the Parties' combined market shares for metallic coated steel for packaging used in non-beverage applications do not materially differ from those for metallic coated steel for packaging as a whole.

684               […].

685               […].

686               […].

687               […].

688               The geographic overlap is essentially limited to Italy. Ilva supplies more than […]% of its steel for packaging to customers located in Italy, where ArcelorMittal generates only about […]% of its steel for packaging sales in the EEA. The overlap in other Southern European countries is very limited. Therefore, the merged entity's market share in other Southern European countries will not be materially affected by the Transaction compared to the pre-Transaction market structure.

689               Confirmed non-confidential minutes of a conference call with a customer on 12.6.2017. Doc ID253.  690  Confirmed non-confidential minutes of a conference call with a customer on 24.7.2017, Doc ID2225.

 691              Confirmed non-confidential minutes of a conference call with a customer on 24.7.2017, Doc ID2225.

692               Confirmed non-confidential minutes of a conference call with a customer on 12.12.2017, Doc ID3741.

693               Replies  to  questions  82,  82.1,  83  and  83.1  of  Q3  –  Questionnaire  to       Customers  (Industrial); Doc ID2800.

694               Combined shares at EEA level of [5-10]% (small welded carbon steel tubes), [10-20]%   (small welded carbon standard tubes), [5-10]% (cold-finished small welded carbon tubes).

695               Ilva is not active in small welded carbon steel precision tubes, nor in hot-finished small  welded carbon steel tubes.

696               See, in general, replies to Q6 – Questionnaire to Tube Customers.

697               Replies  to  question  16  of  Q6  –  Questionnaire  to  Tube  Customers.  See        also  replies  to  Q5  – Questionnaire to Tube Producers.

698               See, for example, replies to question 12 of Q6 – Questionnaire to Tube Customers.

699               Replies to question 16 of Q6 - Questionnaire to Tube Customers.

700               Replies to question 23 of Q5 – Questionnaire to Tube Producers; see also replies to question 17 of  Q6 - Questionnaire to Tube Customers.

701               From time to time Ilva sells limited volumes to other countries from its SSCs in Italy and France. In Slovakia and the Czech Republic the Parties’ combined market shares also reach [20-30]% (2016). However, Ilva’s market shares and the market share increments in these countries are [0-5]% and [0-5]% respectively and those horizontal overlaps are not discussed further in this Decision.

702               Ilva  makes  some  limited  sales  from  its  SSCs  in  Italy  and  France  to  other  countries  where  also ArcelorMittal is present. In the Czech Republic and Slovakia the combined market share also rises       to [20-30]% and [20-30]% respectively (2016). However, the market share increment brought by the transaction is considerably low, less than [0-5]%, in both countries. They are thus not discussed further in this decision.

703               ArcelorMittal  has  suggested  that  the  distribution  of  metallic  coated  steel  for   packaging  through specialised SSCs would constitute a distinct market. However, there is no overlap in this potential segment as Ilva is not present. The volumes concerned are also limited compared to the overall distribution of flat carbon steel through SSCs. The outcome of the assessment would thus remain the same even if the market definition alternative suggested by the Notifying Party was considered.

704               For instance, Ilva only sold […] kt of slabs in 2016 and achieved a market share of [0-5]% in  the EEA. In contrast, ArcelorMittal sold […] kt of slabs in Europe in 2016 and achieved a market share of [5-10]%.

705               For instance, ArcelorMittal has approximately [5-10]% sales market share for slabs in the EEA.

706               The Notifying Party submitted that, to the best of its knowledge, HR is typically not used as a direct input in the production of metallic coated steel for packaging in Europe.

707               The  Notifying Party submitted  that,  to  the  best  of its  knowledge,  GS  is typically not  used    in the production of metallic coated steel for packaging in Europe.

708               Figures are for the UK only. The Parties do not have SCs in Ireland.

709 Form CO, paragraph 750.

710 Form CO, paragraph 42.

711               Tailor welded blanks are manufactured by cutting to size and welding together various individual pieces of uncoated or coated steel sheets such as HR, CR and GS. The Commission has previously   concluded that all tailor welded blanks belong to the same relevant product market and that the market is likely at least EEA-wide though it has left the exact geographic market definition ultimately open. See, for instance, M.5545 – ArcelorMittal/Noble European Holding, paragraphs 14 and 21–2. The outcome of the competitive assessment is not dependent on the exact market definition.

712               Sandwich panels are used in the construction industry for applications such as cladding or roofing. They typically consist of an insulating core that is covered by steel facings. The Commission has previously considered that the markets for the panels could be delineated for instance according to the insulation material (foam or mineral wool) but has left the question ultimately open. The Commission has considered that the market for mineral wool panels is EEA-wide but that for foam panels it could be narrower, leaving the exact market definition open. See, for instance, M.7479 – Kingspan/Steel Partners, paragraphs 21–4 and 59–61.The outcome of the competitive assessment is not dependent on the exact market definition.

713               Steel construction sheets are used in the construction industry for applications such as  roofing, decking and cladding. They are typically produced from GS or OC steel and are usually profiled. The Commission has previously considered that the market may be delineated according to the end-use of the sheets but has ultimately left the exact market definition open. The Commission has previously considered that the markets are national in scope but has left it open whether the market could be cross- border national for certain countries. See, for instance, M.7155 – SSAB/Rautaruukki¸paragraphs 54-6 and 123–4; and M.7479 – Kingspan/Steel Partners, paragraphs 27–8 and 65. The outcome of the competitive assessment is not dependent on exact the market definition.

714               Horizontal Merger Guidelines, paragraph 39.

715               Judgment  in Bertelsmann  and  Sony  Corporation  of America  v Impala, C-413/06 P, EU:C:2008:392, paragraph 122; Horizontal Merger Guidelines, paragraph 39.

716               Horizontal Merger Guidelines, paragraph 40.

717               Judgment of 6 June 2002, Airtours v Commission, T-342/99, EU:T:2002:146, paragraph   62; Judgment in Bertelsmann and Sony Corporation  of  America  v  Impala,  C-413/06  P,  EU:C:2008:392, paragraph 123; Horizontal Merger Guidelines, paragraph 41.

718               Judgment  in Bertelsmann  and  Sony  Corporation  of America  v Impala, C-413/06 P, EU:C:2008:392, paragraph 125.

719               Horizontal Merger Guidelines, paragraph 42.

720               Doc ID1658-47705 […]. Doc ID1469-74220 […].

721               Form CO, paragraph 29, reply to the Article 6(1)(c) decision, section III C.

722               Reply to the SO, Heading IV, paragraphs 327 et seq.

723               Horizontal Merger Guidelines, paragraph 43.

724               Commission decision in Case IV/35.814 – Extras for steel alloys, dated 21 January 1998.

725               Commission decision in Case IV/E-1/35.860-B Seamless steel tubes, dated 8 December 1999.

726               Commission decision in Case COMP/38.344 – Prestressing steel, dated 30 June 2010.

727               Doc ID1658-46391 […].

728               Case Gencor Ltd v Commission, paragraph 276, and Case Airtours v Commission, paragraph 60. See also Horizontal Merger Guidelines, paragraph 44.

729               See Case M.7881 – AB Inbev/SAB Miller, paragraph 52 and subsequent. See also the DOJ complaint  in

U.S. v. Anheuser-Busch InBev SA/NV and Grupo Modelo S.A.B. de C.V., available at https://www.justice.gov/file/486606/download.

730               Reference is made to the seminal work of Rotemberg and Saloner (Rotemberg, J. J., & Saloner, G. (1990). Collusive price leadership. The Journal of Industrial Economics, 93-111), which describes it as follows ‘one of firms announces a price change in advance of the date at which the new price will take effect and the new price and date are swiftly matched by the other firms in the industry’. According to the literature, such coordinating behaviour is conceivable in presence of some kind of frictions among market players, which generally determines the leadership role of one company. Such frictions may include, for example, asymmetric information as described by Rotemberg and Saloner (1990), capacity constraints as described by Ishibashi (Ishibashi, I. (2008). Collusive price leadership with capacity constraints. International Journal of Industrial Organization, 26(3), 704-715), asymmetric costs or differentiated products as described by Mouraviev and Rey (Mouraviev, I., & Rey, P. (2011). Collusion and leadership. International Journal of Industrial Organization, 29(6), 705-717), etc.

731               Reply to the SO, p. 195.

732               See recital (757).

733               See  recital  (756)  et  seq. According to  a  customer,  ‘the  "base  price  +  extra" pricing model locked ArcelorMittal in a price level...Without this model, ArcelorMittal can more easily increase the price of the steel where they believe there is less competition. Reply to question 6.2 of Q12 - Questionnaire 12  to industrial customers (Phase II), Doc ID2898.

734               Replies to question 6.2 of Q12 – Questionnaire to industrial customers (Phase II): ‘Reducing extras from final prices you can get the base price.’; ‘It's not that easy but you can make the reverse engineering and calculate backward a base price’ (Doc ID3225). ArcelorMittal itself appears to converts effective prices into BPEQ (base price equivalent) for internal analytical purposes (see, for example, RFI 23 – Annex 7 – […]).

735               See, for examples, reports listed in reply to RFI 7, Annex 7.2, 7.4 and 7.5.

736               Doc ID1661-1787 […].

737               Doc ID1661-47912 […].

738               For the reasons explained in Section 10.5.4 notably intelligence that suppliers gather concerning their rivals, public price increase announcements, and price indices, the markets for flat carbon steel in        the EEA are transparent.

739               Doc ID272.

740               Doc ID1469-11107 […].

741               Replies to question 74 of Q3 – Questionnaire to customers (industrial), Doc ID2800, and replies to question 35 of Q11 –Questionnaire to industrial customers (Phase II), Doc ID3225.

742               Replies to question 75 of Q3 – Questionnaire to customers (industrial), Doc ID2800.

743               Replies to question 34 of Q11 – Questionnaire to industrial customers (Phase II), Doc ID3225.

744               Doc ID1658-43897 […].

745               […], Doc ID1659-49580 […].

746               Doc ID1660-95596 […].

747               Doc ID555-1259 […].

748               Doc ID1659-98759 […].

749               See, for example, replies to Question 34 of Q11 – Questionnaire to industrial customers (Phase II), Doc ID3225.

750               Doc ID555-1223 […].

751               Doc ID617-1560 […].

752               Doc ID1659-4537 […].

753               Doc ID1658-9257 […].

754               Doc ID1660-39517 […].

755               Doc ID1661-10767 […].

756               Doc ID1469-81021 […].

757               Doc ID1659-60067 […]. See recital (1159) et seq.

758               Doc ID1469-88297 […].

759               Doc ID1661-30069 […]. See Figure 113 and Figure 114 below.

760               Doc ID1660-3097 […].

761               Doc ID1659-98759 […].

762               Doc ID1660-95596 […].

763               Doc ID1661-30069 […].

764               Doc ID1658-65974 […].

765               See also RFI 7, Annex 43-3, […].

766               RFI 7, Annex 43-3, […].

767               RFI 7, Annex 43-3, RFI 43, Annex 7 - […].

768               As shown in Figure 115, […].

769               Figure 113 shows that […].

770               […] Doc ID1660-95596 […].

771               Doc ID1661-30943 […], see also Doc ID1661-67030 […] and Doc ID1661-18641 […].

772               See recitals (1138) and (1154).

773               Horizontal Merger Guidelines, paragraphs 49 and 50.

774               See Section 9.2.

775               Replies to question 30 of Q11 –Questionnaire to industrial customers (Phase II), Doc ID3225.

776               See, for example, Figure 113 and Figure 114 […].

777               […].

778               Replies to questions 70 and 71 of Q3 – Questionnaire to customers (industrial), Doc ID2800.

779               […], Doc ID1659-49580 […].

780               Replies to question 98.1 of Q1 – Questionnaire 1 to competitors (EEA supplier of flat carbon steel), Doc ID2798.

781               Replies to  question 97  of Q1  – Questionnaire  1  to  competitors (EEA supplier  of flat   carbon steel), Doc ID2798.

782               Replies to question 97.1 of Q1 – Questionnaire 1 to competitors (EEA supplier of flat carbon steel), Doc ID2798.

783               See, for example, documents in Annex 43 to RFI7.

784               Annex 43 to RFI7, […].

785               Doc ID1469-87360 […].

786               Reply to the SO, paragraph 343 et seq.

 787              Reply to the SO, paragraph 365 et seq

788               Section 10.5.1.

789               Reply to the SO, paragraph 351 et seq.

790               Horizontal Merger Guidelines, paragraph 52.

791               Horizontal Merger Guidelines, paragraph 53.

792               Horizontal Merger Guidelines, paragraph 54.

793               Horizontal Merger Guidelines, paragraph 55.

794               Doc ID1469-84679 […].

795               Doc ID1658-21449 […].

796               Price for OC comprises without prejudice to any other extras, the base price for HR coil, and  the extras for organic coating).

797               Reply to the SO, paragraph 339.

798               Reply to the SO, paragraphs 376 et seq.

799               See, for example, […].

800               Reply to the SO, paragraphs 379 et seq.

801               Doc ID1659-44486.

802               Doc ID1660-53695 […].

803               Doc ID1661-56721 […].

804               Replies to Question 37 of Q11 -Questionnaire to industrial customers (Phase II), Doc ID3225. 805 Replies to Question 37 of Q11 - Questionnaire to industrial customers (Phase II), Doc ID3225. 806            Replies to Question 37 of Q11 – Questionnaire to industrial customers (Phase II), Doc ID3225.

 807              See, for example, Doc ID1660-53695 […].

808               See recital (1196).

809               Doc ID1469-61892 […].

810               […].

811               Doc ID1469-56546 […].

812               Doc ID1658-11876 […], see also Doc ID1658-81525 […].

813               Doc ID1658-3839 […].

814               For example, […]. Evidence shows that while, during the dumping period, ArcelorMittal   attempted to endorse a price premium […] (see recital (1194)), but at the end of 2015, its price guidance was not sufficiently followed by EEA rivals.

815               For the purpose of calculating the HHI index, imports were not calculated as a single entity but as a group of competitors with 1% each.

816               See Section 11.1.

817               […] (see, for example, Doc ID1469-95072 […], Doc ID1661-79219 […].

818               Doc ID1661-79219 […], p. 13.

819               Doc ID1658-75943 […].

820               Doc ID2454-13 (RFI 23 - Annexes - Agora CW March 2016).

821               Doc ID1659-3114 […].

822               […]. Doc ID2454-485 (RFI 23 - Annexes - 29 2012 ITALY).

823               See also Section 9.4.3.4.

824               Doc ID1659-38775 […].

825               Reply to question 75 of Q3 – Questionnaire to customers (industrial), Doc ID2800.

826               Confirmed non-confidential minutes of a call with a competitor on 4.7.17, Doc ID2281, confirmed non- confidential minutes of a call on 21.8.2017, Doc ID1779, confirmed non-confidential minutes of a call with a customer on 13.6.17, Doc ID1744, confirmed non-confidential minutes of a call with a competitor on 20.7.17, Doc ID423, confirmed non-confidential mintues of a call with a competitor on 18.7.2017 Doc ID443, confirmed non-confidential minutes of a call  with a  customer  on 28.4.2017  Doc ID272.

827               Doc ID1661-79219 […].

828               Doc ID1469-56546 […].

829               See recital (1192).

830               […]. Form CO, p. 2-3.

831               Doc ID1659-28228 […].

832               Doc ID1659-28228 […].

833               Doc ID1661-79219 […]. ‘MCG’ stands for ‘Marcegaglia’.

834               Replies to question 76 of Q3 - Questionnaire to customers (industrial), Doc ID2800.

835               […].

836               Doc ID1661-79219 […].

837               Doc ID1661-79219 […]. […].

838               […].

839               See, for instance, Doc ID1469-6401 […] from 29 January 2017.

840               Doc ID1469-6401 […].

841               Reply to the SO, paragraph 314.

842               Doc ID3294-1200 (Replies + annexes RFI 31 - DOC-000001200.msg).

843               Doc ID1658-34929 […].

844               Doc ID1658-34929 […].

845               […].

846               Doc ID3294-1153 […], DOC-000001153 msg).

847               Doc ID1659-18384 […].

848               Doc ID3294-1161 […], DOC-000001161 msg) […].

849               Doc ID3294-1195 ([…], DOC-000001195 msg).

850               Doc ID1658-34576 […].

851               Doc ID1658-35620 […].

852               See Figure 118.

853               Replies to question 84 of Q3 – Questionnaire to customers (industrial), Doc ID2800 and question  90 of Q4 – Questionnaire to Customers (SSCs), Doc ID2804.

854               Replies to question 84 of Q3 – Questionnaire to customers (industrial), Doc ID2800.

855               Replies to question 90 of Q4 – Questionnaire to customers (SSCs), Doc ID2804.

856               Replies to question 78 of Q3 – Questionnaire to customers (industrial), Doc ID2800.

857               Courtesy  translation  from  Polish:  ‘AM  pozyskując  potencjał  Ilvy  stanie  się głównym dominującym producentem europejskim i będzie mógł dyktować warunki na rynku’.

858               Replies to question 84 of Q4 – Questionnaire to Customers (SSCs), Doc ID2804.

859               Courtesy translation from German: 'Da auch Anti-Dumping auf hot-rolled besteht, und sich nun 2 starke hot-rolled Lieferanten zusammen schließen, liegt dies sehr nahe.'

860               Courtesy translation from Italian: ‘In una situazione di quasi monopolio, il prezzo verrà influenzato dalle decisioni del nuovo soggetto’.

861               Courtesy translation from Italian:    ‘Da metà degli anno '90 Ilva ha avuto una funzione calmierante sul mercato italiano a vantaggio degli utilizzatori finali e delle loro esportazioni. Pertanto l'acquisizione di Ilva da parte di Arcelor induce a pensare ad un futuro aumento di prezzi in Italia…‘.

862               Replies to question 79 of Q3 – Questionnaire to customers (industrial), Doc ID2800 and question  85 of Q4 – Questionnaire to customers (SSCs) Doc ID2804.

863               Replies to  questions 80  and  81 of Q3 – Questionnaire  to  customers (industrial),  Doc      ID2800 and questions 86 and 87 of Q4 – Questionnaire to customers (SSCs), Doc ID2804.

864               Replies to questions 80 and 81 of Q3 – Questionnaire to customers (industrial,) Doc ID2800.

865               Replies to questions 86 and 87 of Q4 – Questionnaire to customers (SSCs) Doc ID2804.

866               Replies to question 17 of Q7 – Questionnaire to customers of SSCs, Doc ID2807.

 867              Replies to question 16 of Q7 – Questionnaire to customers of SSCs, Doc ID2807.

 868              Courtesy translation from French ‘AM a toujours été plus cher que ILVA’.

869               Courtesy translation from French ‘Ilva est la plus grosse forge européenne. Si ArcelorMittal achète Ilva, il aura une position dominante sur le marché européen lui permettant d'appliquer de nouvelles conditions tarifaires’.

870               Courtesy translation from French:  ‘Cette  entreprise  va  être  en  position  quasi monopolistique, nous pouvons craindre une augmentation des prix’.

871               Courtesy translation from Italian: ‘potrebbe portare ad un ulteriore aumento generalizzato dei prezzi del mercato Italiano dell'acciaio, anche perché già ora i prezzi di ARCELOR MITTAL in Europa sono indicativi per tutto il settore’.

872               See replies to question 16 of Questionnaire 7 to customers of SSCs (Doc ID2807).

873               ¸Courtesy  translation  from  Italian:  ‘Perdita  dei  nostri  mercati  esteri  perchè  con  il      prezzo  che aumenteranno non potremmo più fornire i nostri distributori all'estero’.

874               Courtesy translation from French: ‘Nous travaillons en direct avec ArcelorMittal et Ilva. Ilva   avait un positionnement tarifaire intéressant, qui nous permettait de maintenir des niveaux de prix réaliste sur le marché de l'industrie Française. Si ArcelorMittal achète Ilva, sa position dominante sur le marché français lui permettra d'imposer de nouvelle conditions tarifaire. Nous achetons également de  la matière à d'autre SSC qui se fournissent chez Ilva ou Arcelor. Elles seront également impactées et elles nous répercuterons ces hausses de prix. Nous serons donc impactés par cette contraction du nombre de fournisseurs sur le marché Français.’

875               Courtesy translation from French: ‘perte d'un fournisseur (ilva) fiable et compétitif, quasiment   un seul fournisseur possible: arcelor, hausse des prix, puisque arcelor pourra faire les prix qu'il veut, allongement des delais (puisque arcelor est beaucoup moins réactif que ilva)’.

876               Confirmed non-confidential minutes of a call with a customer on 7 December 2017, Doc ID3761.

877               Confirmed non-confidential minutes of a call with a customer on 7 December 2017, Doc ID3658.

 878              Confirmed non-confidential minutes of a call with a customer on 26 April 2017, Doc ID3631.

879               Confirmed non-confidential minutes of a call with a customer on 14 July 2017, Doc ID1850.

880               Confirmed non-confidential minutes of a call with a customer on 4 July 2017, Doc ID2281.

881               Confirmed   non   confidential   minutes   of   a   call   with   an   anonymous   market     participant   on 15 December 2017, Doc ID3574

882               Confirmed non-confidential minutes of a call with a customer on 25 April 2017, Doc ID1946.

883               See, for example, Annex 1.3 to the reply to RFI 20. As discussed in more detail in Section 4.2., the Commission considers that such contacts are likely to affect the reliability of opinions stated by respondents to the Commission's market investigation.

884               All above quotes are from Annex 1.4 to the reply to RFI 20.

885               See Annex 3.4 to the reply to RFI 20.

886               Commission's Notice on Remedies acceptable under Council Regulation (EC) No 139/2004   and under Commission Regulation (EC) No 802/2004 ('Remedies Notice'), OJ C 267, 22.10.2008, p. 1.

887               Remedies Notice, paragraph 5.

888               Remedies Notice, paragraph 9.

889               Remedies Notice, paragraph 9.

890               Remedies Notice, paragraph 12.

891               Remedies Notice, paragraph 9.

892               Remedies Notice, paragraph 11.

893               Remedies Notice, paragraph 22.

894               Remedies Notice, paragraphs 23-25.

895               Remedies Notice, paragraph 26.

896               Remedies Notice, paragraph 32.

897               Remedies Notice, paragraph 47.

898               Including [commercial information].

899               Including [commercial information].

900               Paragraph 101.

901               Replies to question 1 of MT2 – Market Test Customers, Doc ID5119.

902               Replies to question 1 of MT1 – Market Test Competitors, Doc ID5118.

903               Replies to questions 2–4 of MT2 – Market Test Customers, Doc ID5119.  904                   Replies to questions 2–4 of MT1 – Market Test Competitors, Doc ID5118.

905               Replies to question 5 and 6 of MT2 – Market Test Customers, Doc ID5119.

906               Replies to questions 5 and 6 of MT2 – Market Test Competitors, Doc ID5118.

907               Replies to question 11 of MT2 – Market Test Customers, Doc Id5119; and replies to question 11 of MT1 – Market Test Competitors, Doc ID5118.

908               Replies to question 12 of MT2 – Market Test Customers, Doc ID5119; and replies to question 12 of MT1 – Market Test Competitors, Doc ID5118.

909               Replies to question 15 of MT1 – Market Test Competitors, Doc ID5118; and replies to question 15 of MT2 – Market test Customers, Doc ID5119.

910               Replies to questions 25 and 26 of MT1 – Market Test Competitors, Doc ID5118; and replies to questions 22 and 23 of MT2 – Market Test Customers, Doc ID5119.

911               [Commercial information].

912               The GS and HDG capacity share of Ilva are slightly higher than the ones presented in Table 22 of the Decision,  which referred to 2016. This is the result of a limited increase in Ilva's HDG capacity       from 2016 to 2017.

913               Sales share for 2016 are computed as follows. For AM, Ilva, Marcegaglia, imports and market size: actual data provided by the Parties. For competitors the Parties only provided sales data up to September/October 2016. Sales in the missing months have been then attributed to competitors on the basis of their sales during the period January/September 2016. As the Parties sales and the market supply are based on actual data the attribution of sales to competitors has no impact on the combined share of the merging parties.

914               Ibidem.

915               The Notifying Party submitted that […].

916               The Notifying Party submitted that […].

917               Replies to question 9 of MT1 – Market Test Competitors, Doc ID5118; and replies to question 9 of MT2 – Market Test Customers, Doc ID5119.

918               Question 2 of RFI no 37,20 March 2018.

919               Replies to question 9 of MT1 – Market Test Competitors, Doc ID5118; and replies to question 9 of MT2 – Market Test Customers, Doc ID5119.

920               [Commercial information].

921    Replies to question 9 of MT1 – Market Test Competitors, Doc ID5118; and replies to question 9 of MT2 – Market Test Customers, Doc ID5119.

922    Replies to question 9 of MT1 – Market Test Competitors, Doc ID5118; and replies to question 9 of MT2 – Market Test Customers, Doc ID5119.

923               Replies to question 12 of MT1 – Market Test Competitors, Doc ID5118; and replies to question 12 of MT2 – Market Test Customers, Doc ID5119.

924               Replies to question 7 of MT2 – Market Test Customers, Doc ID5119.

925               Replies to question 15 of MT1 – Market Test Competitors, Doc ID5118; and replies to question 15 of MT2 – Markets Test Customers, Doc ID5119.

926               Replies to question 16 of MT2 – Market Test Customers, Doc ID5119.

927               Replies to question 22 of MT2 – Market Test Customers, Doc ID5119.

ANNEX

1                  Letter of Facts, section 1.19.

2                  Contrary to the Notifying Party's arguments present in the Reply to the SO, the aim of the Commission's quantitative analysis is not to measure the extent by which non-EEA exporters would be willing to supply more to EEA customers, which is likely not the binding constraint to imports to the EEA.

3                  Reply to the Letter of Facts, p. 8.

4                  In contrast, the "cross price elasticity between the domestic price and customer demand for imported HR proposed by the Notifying Party does not provide a direct measure of this constraint.

5                  Endogenous events, that is events under the control of a company or perfectly foreseeable,   as opposed to  exogenous  events,  may  introduce  an  additional  complexity  in  the  model  in  relation  to       the identification of the effects of such events. Endogeneity issues are further discussed in Section 5.1 of this Annex.

6                  See Commission, AD630 investigation and Regulation 2017/649.

7                  See Commission, AD635 investigation and Regulation 2017/1795.

8                  A registration requirement implies that, in the event of final anti-dumping measures, tariffs would be applied retroactively to the imported volumes from the day the registration requirement was issued.

9                  Domestic HR market prices in China are taken as a proxy of the domestic prices from exporting countries.

10                The third lag means that at period t, the relative price series of t-3 is plotted. That is, it is the relative price of three months earlier.

11                A  three  months  delay  between order and delivery  appears  to  be  appropriate, based on  the evidence gathered during the Commission's  market  investigation.  See,  for  example,  recitals  (582)-(583)  of the Decision.

12                In a few occasions information was missing for a given month. To construct a price series without gaps, the Commission filled the missing months with the simple average of the month before and the month after.

13                 S&P Global Platts.

14                ID3795 provides the details of the methodology followed by SBB to compute the price series. It  is also available at: https://www.ptts.co/IM.Platts.Cotent/MethodologyReferences/MethodologySpecs/steel.pdf.

15                 The "first difference" is the difference in the values of a variable at time t and at time t-1.

16                A dummy variable is binary variable that takes values zero and one, and is generally associated   with a given period or a given group.

17                This is because across the four regressions the shares sum to 100% at any given point in time.

18                In its Reply to the SO (Annex 3.B.1), the Notifying Party argued that the use of dummy variables to identify the Ilva Caps is not flexible and leads to imprecise estimates. For this reason, it suggested to use a continuous variable that only takes four discrete values reflecting Ilva's capacity reductions at  three points in time (i.e. following the mothballing of BF1, the mothballing of BF5 and the restart of BF1). The Commission disagrees with this approach. First, contrary to the Notifying Party's argument, the use of two distinct dummy variables is a more flexible approach, as the methodology proposed by the Notifying Party constraints the effect of the imposition of the 2012 and 2015 Ilva Cap to one estimated value. Second, the alleged ability of this approach to disentangle the effect of the 2015 Ilva Cap and the dumping period is questionable. It is based on the estimation of a single coefficient on a variable which takes only four values over time, one of which (when both BF1 and BF5 were not operative) lasting only for a few months, and the timing of another change being near simultaneous to the dumping. In other words, the variable used by the Notifying Party is very similar to a constrained version of the two dummy variables used by the Commission. As such, it is affected by the same limitations in correctly identifying the effects of the further reduction of Ilva's capacity and of the dumping period.

19                 Reply to the SO, Annex 3.B.1.

20                The exclusion of those dummy variables is the main factor introduced by the Notifying Party   affecting the analysis of the volume diversion following the imposition of the Ilva Cap, and resulting in a significant diversion of sales shares to imports.

21                The Notifying Party also argued that it is not appropriate to include price variables as explanatory variables because they might be affected by the imposition of the Ilva Cap. In this respect, the Commission notes that, in the Notifying Party's specification without prices, the results do not materially change. Moreover, as explained in Section 3 of this Annex, it is important to properly control for the availability of imports in the EEA, as to not erroneously attribute changes in sales shares to the imposition of the Ilva Cap.

22                A complete discussion of the details of endogeneity and instrumental variable regression can  be found, for example, in Wooldridge (2002) Econometric analysis of Cross Section and Panel Data or Hayashi (2000) Econometrics.

23                Contrary to the Notifying Party's argument in its Reply to the SO, in the Commission's view the HR price in China can be treated as an exogenous variable. This is because, while the Chinese HR price is likely to affect the availability of imports in Southern Europe, the latter is unlikely to have any appreciable effect on the HR price in China because the size of the HR market in China is several times larger than the HR volumes imported in Southern Europe. Hence, there is no simultaneous causality problem. Moreover, the omitted variable bias referred to by the Notifying Party (that is, the domestic price in other exporting countries not being included in the regressions) is ruled out by assumption because, as explained in footnote 9, the Commission's uses the HR price in China as a proxy of the domestic prices from exporting countries. For all practical purposes it is not possible to include prices in all possible exporting countries in the analysis. The Commission considers that, for the purposes of the quantitative analysis, it is reasonable to use the HR price in China as a proxy for the conditions at which HR imports are available to EEA customers. This approach also seems to be in line with the Notifying Party's emphasis on the importance of China as a driver of global steel dynamics (see, for example, Reply to the SO, paragraph 241).

24                The test is an heteroscedasticity and auto-correlation robust version of the Hausman test, it is applied using the endog() option of the command ivreg2 in Stata.

25                Technically,  the  test  failed  to  reject  the  null  hypothesis  of  exogeneity  of  the  domestic  Southern European HR market price.

26                In the regressions, Southern Europe includes Italy, Spain, Portugal and Greece. France is not included because the data do not allow a split in Northern and Southern France. However, the inclusion of France does not significantly change the results.

27                Note  that  ArcelorMittal's  sales  shares  gains  are  higher  than  Ilva's  sales  shares  losses because the coefficients are constrained to sum up to zero.

28                The K-P (Kleibergen-Paap) and  the  S-W  (Sanderson-Windmeijer)  underidentification  tests   are two HAC-robust tests investigating the relevance of the instruments (they are also called rank tests). The  null hypothesis is that the equation is underidentified or not identified. Not rejecting the null hypothesis may cast doubts on the relevance of the instruments.

29                The H-S (Hansen-Sargan) overidentification test is a HAC-robust test investigating the joint  validity of the instruments. The null hypothesis is that the instruments are jointly valid. A rejection of the null hypothesis may cast doubt on the validity of at least some instruments.

30                The S-W (Sanderson-Windmeijer) F statistic is a HAC-robust joint statistic on the instruments that provide insights on whether the instruments are weak. In general, an F statistic above 10 or 20 is considered to indicate that the instruments are not weak. It should be noted, however, that the critical values for this statistic have been tabulated only for the case of i.i.d. error terms. In any event, a relatively high F statistic still suggests that the instruments are not weak.

31                That is, the test rejects the null hypothesis of exogeneity.

32                 That is, the sum of the sales shares of ArcelorMittal, Ilva and Other EEA Suppliers.

33                The residual demand of domestic EEA suppliers is equal to the product of the aggregate demand and the share of demand of domestic producers. The elasticity of the residual demand is hence the sum of the aggregate demand elasticity and the elasticity of the share of demand going to domestic producers.

34                This is calculated  as the ratio  between  the coefficient (i.e.  […]) and the average  sales  share  (i.e. on average […]% in Southern Europe).

35                ID4296 – "Hot-Rolled Flat-Rolled Carbon-Quality Steel Products from Brazil, Japan, and Russia", June 2011, U.S. International Trade Commission. Available at: https://www.usitc.gov/publications/701_731/pub4237.pdf.

36                To the Commission knowledge, there are no other estimates of aggregate elasticity for HR products.

37                 Multi-collinearity arises when two or more variables display a high degree of correlation. The higher the correlation, the more difficult is to accurately estimate the coefficients of those variables. In general, the estimated coefficients of multi-collinear variables have a high absolute value, but are not  statistically significant.

38                Similar issues affect the additional analysis submitted by the Notifying Party in its Reply to the Letter of Facts, attempting to estimate the demand for imports of EEA customers. Due to the inclusion of both domestic and imports' prices in Southern Europe, the model is underidentified and the instruments are weak. Moreover, the model excludes several control variables used in the Commission's analysis,  among which the variables controlling for the availability of imports in the EEA. The reintroduction of those variables renders the coefficients on the price variables non statistically significant.

39                The Commission considers that, in principle, it would be appropriate to use prices that are directly available to customers. However, due to the multi-collinearity issue that arises by introducing imports' prices in Southern Europe in the regressions, this is not possible in this case.