EUROPEAN COMMISSION DG Competition

CASE M.7878 – HEIDELBERGCEMENT/SCHWENK/CEMEX HUNGARY/CEMEX CROATIA

(Only the English text is authentic)

MERGER PROCEDURE

REGULATION (EC) 139/2004

Article 8(3) Regulation (EC) 139/2004

Date: 5.4.2017

This text is made available for information purposes only. A summary of this decision is published in all EU languages in the Official Journal of the European Union.

Parts of this text have been edited to ensure that confidential information is not disclosed; those parts are enclosed in square brackets.

EUROPEAN COMMISSION

Brussels, 5.4.2017 C(2017) 1650 final

Public version

COMMISSION DECISION

of 5.4.2017

declaring a concentration to be incompatible with the internal market and the functioning of the EEA Agreement

(Case M.7878 – HeidelbergCement/Schwenk/Cemex Hungary/Cemex Croatia)

(Only the English text is authentic)

1

TABLE OF CONTENTS

1. Introduction .................................................................................................................. 4

2. The parties and the concentration ................................................................................ 7

3. Union Dimension ......................................................................................................... 9

3.1. Principles ...................................................................................................................... 9

3.2. Factual background .................................................................................................... 10

3.3. The Parties' submissions ............................................................................................ 11

3.4. The Commission's assessment ................................................................................... 12

3.4.1. Initiation of the Transaction ....................................................................................... 12

3.4.2. Organisation of the Transaction ................................................................................. 13

3.4.3. Financing of the Transaction ...................................................................................... 15

3.4.4. Schwenk's involvement in the Transaction ................................................................ 17

3.4.5. DDC's involvement in the Transaction ...................................................................... 18

3.4.6. Conclusion.................................................................................................................. 20

3.5. Partial referral to Hungary ......................................................................................... 20

4. The Investigation of the substance of the Transaction ............................................... 20

5. The Parties' activities in Croatia................................................................................. 21

6. Relevant markets ........................................................................................................ 22

6.1. Description of the products as well as production and distribution assets ................. 22

6.2. Product market definitions ......................................................................................... 24

6.2.1. Previous Commission Decisions ................................................................................ 24

6.2.2. The Notifying Parties' submissions ............................................................................ 24

6.2.3. The Commission's assessment ................................................................................... 24

6.3. Geographic market definition .................................................................................... 27

6.3.1. Previous Commission decisions ................................................................................. 27

6.3.2. The Notifying Parties' submissions ............................................................................ 28

6.3.3. The Commission's assessment ................................................................................... 28

7. Competitive assessment ............................................................................................. 35

7.1. Principles .................................................................................................................... 35

7.2. Application to this case .............................................................................................. 37

7.3. Main suppliers of grey cement in the relevant catchment areas ................................ 38

7.4. Geographic differentiation within the relevant catchment areas ................................ 40

7.5. Market shares ............................................................................................................. 42

7.5.1. Relevance for the competitive assessment of market shares and the number of alternative suppliers of grey cement .......................................................................... 42

7.5.2. Market share methodology......................................................................................... 45

7.5.3. Sales market shares .................................................................................................... 48

2

7.5.4. Capacity market shares .............................................................................................. 53

7.5.5. Market shares and geographic differentiation ............................................................ 55

7.6. Competitive pressure exerted by DDC and Italcementi on Cemex before the Transaction ................................................................................................................. 59

7.6.1. DDC is an important competitive force in Croatia .................................................... 59

7.6.2. DDC is a close competitor of Cemex Croatia ............................................................ 60

7.6.3. Italcementi is an important competitive force in western Croatia ............................. 63

7.7. Constraints from competing suppliers of grey cement .............................................. 63

7.7.1. General factors limiting the constraints exerted by competing land-based and sea-borne suppliers of grey cement .................................................................................. 63

7.7.2. Constraints exerted by specific individual land-based competitors suppliers of grey cement ........................................................................................................................ 66

7.7.3. Competitive constraints exerted by other potential land-based importers ................. 78

7.7.4. Factors limiting the constraints exerted specifically by competing sea-borne suppliers of grey cement ............................................................................................................ 79

7.7.5. Cumulative constraints from competitors .................................................................. 82

7.8. Spare capacity and incentives to expand of competing suppliers of grey cement ..... 82

7.9. Likelihood of entry by other suppliers of grey cement .............................................. 85

7.10. Countervailing buyer power ....................................................................................... 88

7.11. Threat of actions by the merged entity further deterring entry or expansion by competing suppliers of grey cement .......................................................................... 89

7.12. Impact of the Transaction........................................................................................... 93

7.12.1. Internal projections by DDC ...................................................................................... 93

7.12.2. The Parties' analysis of price effects .......................................................................... 96

7.12.3. Feedback from customers and competitors ................................................................ 99

8. Significant impediment of effective competition in a substantial part of the internal market ....................................................................................................................... 101

9. Commitments ........................................................................................................... 102

9.1. Principles .................................................................................................................. 102

9.2. The Commitments submitted by the Notifying Parties and the negotiations with Titan and Asamer............................................................................................................... 103

9.3. The Parties' submissions .......................................................................................... 106

9.4. Responses to the market test of the Commitments .................................................. 107

9.5. The Commission's assessment ................................................................................. 108

9.5.1. Structural deficiencies of the Commitments ............................................................ 108

9.5.2. Low likelihood of finding a suitable lessee.............................................................. 110

9.5.3. Concerns about the scale of the Commitments ........................................................ 119

9.5.4. Concerns regarding the implementation modalities of the Commitments ............... 127

10. Conclusion and compatibility with the internal market ........................................... 127

COMMISSION DECISION

of 5.4.2017

declaring a concentration to be incompatible with the internal market and the functioning of the EEA Agreement

(Case M.7878 – HeidelbergCement/Schwenk/Cemex Hungary/Cemex Croatia)

(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, and in particular Article 8(3) thereof,

Having regard to the Commission's decision of 10 October 2016 to initiate proceedings in this case,

Having given the undertakings concerned the opportunity to make known their views on the objections raised by the Commission,

Having regard to the opinion of the Advisory Committee on Concentrations,

Having regard to the final report of the Hearing Officer in this case,

1. INTRODUCTION

(1) On 5 September 2016 the Commission received a notification of a proposed concentration pursuant to Article 4 of Council Regulation (EC) No 139/2004 (the "Merger Regulation") by which the undertakings HeidelbergCement AG (“HeidelbergCement”, Germany) and Schwenk Zement KG ("Schwenk", Germany) acquire, through their jointly controlled joint venture company Duna-Dráva Cement Kft. ("DDC"), joint control within the meaning of Article 3(1)(b) of the Merger Regulation of the whole of the undertakings Cemex Hungária Építőanyagok Kft ("Cemex Hungary") and Cemex Hrvatska dd ("Cemex Croatia"), both part of Cemex, S.A.B. de C.V ("Cemex Group"), by way of purchase of shares ("the Transaction").

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 337, 14.9.2016, p. 7.

5 OJ L 24, 29.1.2004, p. 1 (the "Merger Regulation").

(2) For the purposes of this Decision, HeidelbergCement, Schwenk, DDC, Cemex Hungary and Cemex Croatia are jointly referred to as the "the Parties". HeidelbergCement and Schwenk are referred to as "the Notifying Parties".

(3) On 10 October 2016, the Commission raised serious doubts as to the compatibility of the Transaction with the internal market and initiated proceedings pursuant to Article 6(1)(c) of the Merger Regulation in relation to the potential market for grey cement in the catchment areas around the plant of DDC in Kakanj/Bosnia-Herzegovina and of Cemex Croatia in Split/Croatia, in the catchment areas around groups of customers, or in regions in Croatia ("the 6(1)(c) Decision").

(4) On 12 October 2016, the Commission provided non-confidential versions of certain key submissions by third parties collected during the initial investigation to HeidelbergCement and to Cemex. On 11 November 2016, the Commission provided the same key submissions by third parties to Schwenk.

(5) On 20 October 2016, HeidelbergCement and Cemex both submitted written comments on the 6(1)(c) Decision. Schwenk did not submit any comments on the 6(1)(c) Decision. A formal state-of-play meeting took place between HeidelbergCement, Cemex and the Commission on 25 October 2016. Schwenk declined the offer of attending a state-of-play meeting after the initiation of proceedings.

(6) On 6 December 2016, the Commission informed the Parties of the preliminary conclusions of the in-depth investigation conducted in the course of the proceedings during a formal state-of-play meeting.

(7) On 12 December 2016, the Commission issued a Statement of Objections ("the Statement of Objections"). In the Statement of Objections, the Commission reached the preliminary conclusion that the Transaction will significantly impede effective competition in a substantial part of the internal market within the meaning of Article 2(3) of the Merger Regulation.

(8) Access to file was first provided on 13 December 2016. Subsequent access to the file was provided on 26 January 2017. Access to confidential data and information relied upon by the Commission in the Statement of Objections was granted to the Notifying Parties' economic advisors in accordance with the data room procedure on 14

6 The 6(1)(c) Decision was sent to HeidelbergCement and Schwenk. HeidelbergCement, Schwenk and Cemex submitted respectively replies to the 6(1)(c) Decision.

7 ID1146, email of Schwenk to the Commission of 17 November 2016.

8 Cemex confirmed by email of 25 November 2016 to the Commission (ID1594) that Cemex would not formally request access to the file and would directly liaise with the Notifying Parties' external counsel to obtain a copy of the Statement of Objections with a view to submitting observations. Accordingly, the Commission sent the Statement of Objections to the Notifying Parties only. Cemex submitted its reply to the Statement of Objections on 2 January 2017.

9 Only HeidelbergCement retained economic advisors who accessed the data room on 14 and 15 December 2016.

10 Business secrets and other confidential information of third parties within the meaning of Article 339 TFEU, Article 18(3) of the Merger Regulation and Article 17(3) of the Commission Implementing Regulation (EU) No 1269/2013 of 5 December 2013 amending Regulation (EC) No 802/2004 implementing Council Regulation (EC) No 139/2004 on the control of concentrations between undertakings (the "Implementing Regulation", OJ L 336, 14.12.2013, page 1) can exceptionally be made available to the addressee of a Statement of Objections within the framework of the data room procedure and under the strict conditions set out in data room rules. The data room procedures are set in the Best Practices on the disclosure of information in data rooms, 2 June 2015, available at http://ec.europa.eu/competition/mergers/legislation/disclosure information data rooms en.pdf.

and 15 December 2016. On the same days, access to confidential third party submissions relied upon by the Commission in the Statement of Objections was granted to the Notifying Parties' legal advisors in accordance with the data room procedure.

(9) On 3 January 2017, both HeidelbergCement and Schwenk replied to the Statement of Objections.

(10) On 11 January 2017, an oral hearing was held.

(11) On 18 January 2017, the Commission adopted a decision pursuant to Article 10(3) second subparagraph, third sentence of the Merger Regulation extending the deadline by five working days in agreement with the Notifying Parties in order to allow the Commission to assess the effects of the Transaction on the markets for which there was a risk that effective competition would be significantly impeded.

(12) On 20 January 2017, a state-of-play meeting between the Commission and the Parties took place.

(13) On 25 January 2017, the Commission issued a letter of facts ("Letter of Facts") informing the Notifying Parties about pre-existing evidence that was not relied on in the Statement of Objections, but which, on further analysis of the file, may be relevant to support the Commission's preliminary conclusions. The Letter of Facts also informed the Notifying Parties about certain additional evidence brought to the Commission's attention after the adoption of the Statement of Objections.

(14) Access to the calculations underlying some of the evidence in the Letter of Facts was granted to the Notifying Parties on 26 January 2017.

(15) On 26 January 2017 (day 65 of the second phase procedure), the Notifying Parties offered commitments pursuant to Article 8(2) of the Merger Regulation in order to dispel the significant impediment to effective competition raised by the Transaction. The offer of commitments triggered the automatic extension of the time limit for adopting a final decision by 15 working days pursuant to the first subparagraph, last sentence, of Article 10(3) of the Merger Regulation.

(16) On 1 and 3 February 2017, HeidelbergCement replied to the Letter of Facts ("Reply to Letter of Facts"). On 2 February 2017, Schwenk replied to the Letter of Facts.

(17) On 14 February 2017, the Commission adopted a decision pursuant to Article 10(3) second subparagraph, third sentence of the Merger Regulation extending the deadline by fifteen working days in agreement with the Notifying Parties in order to allow the Commission to assess the effects of the Transaction on the markets for which there was a risk that effective competition would be significantly impeded.

(18) On 20 February, 28 February and 16 March 2017, further access to file was provided to the Notifying Parties

(19) On 24 February 2017, a state-of-play meeting took place between the Commission and HeidelbergCement, Schwenk and Cemex.

(20) On 21 March 2017, the Advisory Committee discussed a preliminary draft of this Decision and delivered a favourable opinion.

11 Only HeidelbergCement's legal advisors chose to access the data room.

12 The Letter of Facts was sent to HeidelbergCement and Schwenk. HeidelbergCement, Schwenk and Cemex submitted respectively replies to the Letter of Facts.

13 Cemex replied to the Letter of Facts of 25 January 2017 on 31 January 2017, ID2759.

(21) On 22 March 2017, further access to file was provided to the Notifying Parties.

(22) On 27 March 2017, the Notifying Parties submitted observations on the documents that were made available to them on 16 and 22 March 2017.

(23) On 27 March 2017, and because further access to file had been exceptionally provided to the Notifying Parties after the meeting of the Advisory Committee on 21 March 2017, the Commission sent the Advisory Committee a further preliminary draft of this Decision. It also informed the competent authorities of the Member States of the possibility to request that a second meeting of the Advisory Committee be convened. None of the competent authorities of the Member States, however, made use of that possibility.

(24) The Hearing Officer provided its favourable opinion on the proceedings in his report which was submitted on 30 March 2017.

2. THE PARTIES AND THE CONCENTRATION

(25) HeidelbergCement is a German producer and distributor of cement, clinker, ready-mix concrete (also referred to as "RMX"), aggregates and other related products. It carries out its commercial activities globally in more than 40 countries and it has several subsidiaries in the territory of the European Economic Area ("EEA").

(26) Schwenk is a family-held limited partnership active in the production of cement, clinker, ready-mix concrete, concrete products and services, as well as aggregates. Its business is focused on Germany, which accounts for more than […]% of its turnover, but it also has limited activities in Austria, France, the Netherlands, Poland and the Czech Republic.

(27) DDC is a full-function joint venture company equally owned and jointly controlled by HeidelbergCement and Schwenk, which is active in Hungary, Croatia and in parts of the Western Balkans (i.e. Bosnia-Herzegovina, Serbia, Macedonia, Montenegro and Albania, all together referred to as "the Broader Region") in the areas of cement, ready-mix concrete and aggregates. In Hungary, DDC currently operates two cement plants and 24 fixed ready mixed concrete plants and two aggregates sites. In the Broader Region, DDC operates one cement plant and 11 ready-mix concrete plants.

(28) Cemex group is a global building materials company headquartered in Mexico active in cement, ready-mix concrete, aggregates and related building materials. It has operations in Africa, the Americas, Asia, Europe and the Middle East. Along with the rest of the Cemex companies, the group will hereinafter be referred to as "Cemex".

(29) Cemex Hungary is a limited liability company under Hungarian law, which is mainly active in the production and sale of ready-mix concrete, paving stones, and aggregates which are only sold in Hungary. Cemex Hungary currently operates 26 fixed ready-mix concrete plants, three aggregate sites and two paving stone plants in Hungary. Cemex Hungary is not active in the production and sale of cement.

(30) Cemex Croatia is a joint stock company under Croatian law, which is mainly active in the production and distribution of grey cement, ready-mix concrete, clinker, and aggregates. Besides its exports to North Africa and the Middle East, Cemex Croatia's activities mainly focus on Croatia and the Broader Region but it also supplies to Italy, Slovenia and Malta. The company operates seven ready-mix concrete plants, five in Croatia, two in Bosnia-Herzegovina and two aggregates quarries in Croatia.

(31) In April 2015, Cemex group initiated a process for the sale of (i) its subsidiary Cemex Austria AG ("Cemex Austria"), which held 100% of the shares in Cemex Hungary, and (ii) its subsidiary Cemex Croatia. Cemex had an interest in selling its subsidiaries together and at the same time.

(32) On 27 July 2015, and against the background of a framework agreement entered into between DDC and Rohrdorfer Baustoffe Austria AG ("Rohrdorfer") on the same day, HeidelbergCement submitted a joint bid to Cemex on behalf of DDC and Rohrdorfer which was retained as the preferred bid.

(33) On 11 August 2015, DDC and Cemex entered into a share purchase agreement pursuant to which DDC will acquire 100% shares in Cemex Croatia. In parallel, Rohrdorfer entered into a similar agreement for the acquisition of Cemex Austria and its subsidiary Cemex Hungary. The acquisition of Cemex Austria and its subsidiary Cemex Hungary by Rohrdorfer was implemented on 2 November 2015. On the same day and following what had been agreed in the framework agreement between DDC and Rohrdorfer, DDC and Cemex Austria entered into a share purchase agreement for the acquisition of Cemex Hungary by DDC.

(34) The acquisition of Cemex Croatia and Cemex Hungary by DDC should be considered as a single concentration within the meaning of the Merger Regulation. HeidelbergCement and Schwenk were interested in acquiring both Cemex Hungary and Cemex Croatia through DDC, while not interested in acquiring Cemex Austria which was to be retained by Rohrdorfer.

(35) Since Cemex was unwilling to spin-off Cemex Hungary from Cemex Austria, DDC and Rohrdorfer entered into their framework agreement according to which they would sign a share purchase agreement for the sale of Cemex Hungary from Rohrdorfer to DDC immediately upon closing of the acquisition of Cemex Austria by Rohrdorfer. The economic reality of the agreements is therefore that DDC acquires control of Cemex Hungary and Cemex Croatia while Rohrdorfer acquires control of Cemex Austria. The acquisitions of Cemex Hungary and Cemex Croatia by DDC were pursued at the same time and are linked through the framework agreement between Rohrdorfer and DDC.

(36) The Transaction thus involves the acquisition of joint control of Cemex Hungary and Cemex Croatia by HeidelbergCement and Schwenk (through DDC) by means of purchase of shares. The Transaction constitutes therefore a concentration within the meaning of Article 3(1)(b) of the Merger Regulation.

16 Merger control clearance by the Austrian Federal Competition Authority was obtained on 19 October 2015.

17 See the Commission’s Consolidated Jurisdictional Notice, OJ C 37, 30.01.2016, p. 3 (the "Jurisdictional Notice"), paragraphs 43 and 47.

3. UNION DIMENSION

3.1. Principles

(37) Pursuant to Article 1(2) of the Merger Regulation a concentration has a Union dimension where the turnover thresholds are met by the "undertakings concerned".

(38) The Merger Regulation does not, however, contain any definition of "undertakings concerned", nor does it prescribe a particular approach as to how such undertakings should be identified.

(39) Such an approach is, however, set out in paragraphs 129 and 145 to 147 of the Consolidated Jurisdictional Notice on the control of concentrations between undertakings ("the CJN").

(40) Paragraph 129 of the CJN states that "[f]rom the point of view of determining jurisdiction, the undertakings concerned are those participating in a concentration […]." In case of a joint-venture acquiring control of another company, it has to be determined whether the joint venture or each of its parent companies should be regarded as the undertaking concerned. In principle the joint venture participating directly in the acquisition of control should be regarded as the undertaking concerned. However, the CJN stipulates in paragraph 145 that the Commission will look at the economic reality of the concentration in order to determine who the real players behind an operation are. In line with that principle, paragraphs 146 and 147 of the CJN specify as follows:

(41) The approach in paragraphs 145 to 147 of the CJN reflects the primary purpose of the Merger Regulation, namely to ensure the effectiveness of merger control and legal certainty for the undertakings involved.

(42) On the one hand, through the approach set out in paragraphs 145 to 147 of the CJN, the Commission seeks to identify, in accordance with the circumstances of fact and of law specific to each case and with a concern to ascertain the economic reality underlying the concentration, whether the parents of a joint venture are in fact the undertakings that are the real players behind the concentration.

(43) On the other hand, the approach in paragraphs 145 to 147 of the CJN reflects the concern for legal certainty by establishing an objective test based on evidence necessarily known to the joint venture and its parents, thereby allowing those undertakings to make an assessment of a concentration’s dimension and to determine the authority or authorities that should be notified of it.

(44) It follows that the Commission is entitled to consider the parent companies of a joint venture as "undertakings concerned" within the meaning of Article 1(2) of the Merger Regulation where the economic reality demonstrates that the parent companies are the real players behind a concentration.

3.2. Factual background

(45) On 20 August 2015, DDC initiated a consultation process with DG Competition regarding the question which would be the "undertakings concerned" by the acquisition of Cemex Hungary and Cemex Croatia for the purpose of the Merger Regulation.

(46) On 28 August and 28 September 2015, DG Competition asked DDC and HeidelbergCement to clarify certain issues and to provide evidence supporting certain statements. DDC and HeidelbergCement provided such clarifications and evidence.

(47) On 13 November 2015, DG Competition informed HeidelbergCement of its opinion that the "undertakings concerned" by the acquisition of Cemex Hungary and Cemex Croatia for the purpose of the Merger Regulation would be HeidelbergCement, Schwenk, Cemex Hungary and Cemex Croatia ("the letter of 13 November 2015"). DG Competition did not receive any response from HeidelbergCement (or Schwenk or DDC) to the letter of 13 November 2015.

(48) In March 2016, HeidelbergCement and Schwenk entered in pre-notification discussions with DG Competition. At no point during such discussions did HeidelbergCement or Schwenk seek in their written submissions to call into question the opinion expressed by DG Competition in the letter of 13 November 2015.

(49) On the contrary, on 25 May 2016, HeidelbergCement and Schwenk requested that the Transaction be examined by the competent authorities of Hungary pursuant to Article 4(4) of the Merger Regulation as regards its effects on Hungarian markets by way of a partial referral to the Hungarian Competition Authority.

(50) Similarly, on 5 September 2016, HeidelbergCement and Schwenk notified the Transaction to the Commission and simply noted that "The European Commission in the consultation process C.1472 – Duna Drava Cement/Cemex Croatia/Cemex Hungary (letter dated November 13, 2015, cf. Annex 1.1.- (1)) held that due to the exceptional and specific circumstances of the Transaction both HC and SCHWENK are not only controlling shareholders of the direct acquirer DDC, whose turnover figures have to be attributed to DDC in order to establish the jurisdiction of the European Commission, but parties to the concentration themselves and thus "undertakings concerned" (see in particular paras. 145-147 CJN)" without contesting the facts underlying the Commission's opinion that HeidelbergCement and Schwenk were the "undertakings concerned" as outlined in the Article 4(4) decision of 22 June 2016.

(51) Moreover, on 27 September 2016, a state-of-play meeting took place during which no attempt was made by HeidelbergCement to call into question the opinion expressed by DG Competition in the letter of 13 November 2015 and in the Article 4(4) decision of 22 June 2016.

3.3. The Parties' submissions

(52) HeidelbergCement argued in its reply to the 6(1)(c) Decision, in its reply to the Statement of Objections and in its reply to the Letter of Facts that, on the acquirer's side, the undertaking concerned by the acquisition of Cemex Croatia is DDC and not HeidelbergCement and Schwenk. Schwenk expressed the same view in its reply to the Statement of Objections and in its reply to the Letter of Facts. In this context, HeidelbergCement and Schwenk brought forward in particular the following reasons:

(53) First, DDC was not set up as a vehicle to enable HeidelbergCement and Schwenk to control the Croatian cement market.

(54) Second, the Transaction was driven by DDC and is meant to benefit first and foremost the interest of DDC to extend its existing business in Croatia and Bosnia.

(55) Third, DDC initiated the Transaction and is the sole party to the respective agreements and bears the risks and opportunities from the Transaction. It cannot be inferred from the fact that HeidelbergCement's support included the arrangement of a joint bid between DDC and Rohrdorfer due to the specifics of the seller’s requirements that HeidelbergCement was the driver of that transaction.

(56) Fourth, HeidelbergCement and Schwenk are affected by the Transaction only as shareholders of DDC. For transactions of the size and importance of the Transaction, DDC seeks support either from local advisors and law firms or from one of its parent companies as it has only a small administration. In addition, it is ordinary business practice that transactions of the size and importance of the Transaction are discussed with the parent companies of DDC.

(57) Fifth, Schwenk is any case not an undertaking concerned because its only involvement concerns the funding of a limited part (roughly […]%) of the Transaction.

3.4. The Commission's assessment

(58) For the reasons set out in recitals (59) to (114), the Commission concludes that because of their significant involvement in the initiation, organisation and financing of the Transaction, HeidelbergCement and Schwenk are the real players behind the Transaction and thus "undertakings concerned".

3.4.1. Initiation of the Transaction

(59) The Transaction was initiated by HeidelbergCement and Schwenk, which identified the Transaction as an attractive business opportunity and decided that DDC should be the acquiring entity. That is confirmed by a number of internal documents submitted by the Notifying Parties and DDC.

(60) First, non-disclosure agreements ("NDAs") were signed by HeidelbergCement and Citigroup representing Cemex on 29 April 2015.

(61) Second, high-level contacts between HeidelbergCement and Cemex were initiated at the latest on 5 May 2015 when a teaser was sent by Citibank/Bank of America representing Cemex in the sales process to [name] Director Group Strategy & Development of HeidelbergCement.

(62) Third, an initial discussion concerning the Transaction took place between HeidelbergCement and Cemex representative [name] on 6 May 2015. HeidelbergCement's impression after that conversation was that it was the preferred purchaser due to its [content of internal documents].

(63) Fourth, it was decided at a meeting on 6 May 2015 attended exclusively by HeidelbergCement's employees ([names]) that "there shall be a steering committee consisting of [names] and chaired by a HeidelbergCement employee, [name].

(64) Fifth, it was decided at the same meeting on 6 May 2015 that [name of HeidelbergCement employee] would be project manager for the Transaction, with the support of two additional HeidelbergCement employees ([names]). The project manager has overall responsibility for the planning and execution of the Transaction.

(65) Sixth, before the meeting of [names] on 6 May 2015, HeidelbergCement and Schwenk agreed to pursue the acquisition of Cemex Hungary and Cemex Croatia with DDC as ultimate purchaser. In this context, HeidelbergCement contacted

ID1088-2267, e-mail exchange between […] and […] dated 29 April 2015.

ID1088-2803, e-mail exchange between […] and […] of 05 May 2015. In that email, […] reports that he believes that the teaser has already been sent to […]. ID437-93, M.7878_Form CO_Annex 5.4.-I.-(01)_Meeting_Minutes_Kickoff.

ID1088-2796, Interner Brief an […] vom 06.05.2015.

ID1088-2796, Interner Brief an […] vom 06.05.2015, [content of internal documents] (courtesy translation of the Commission: [content of internal documents].

(66) Seventh, at the latest on 6 May 2015, HeidelbergCement decided it would submit an indicative offer: [content of internal document].

(67) Eighth, on 7 May 2015, HeidelbergCement informed DDC about the various decisions HeidelbergCement had taken regarding the Transaction, including HeidelbergCement's decisions: (i) to establish a Steering Committee which (ii) should be chaired by a HeidelbergCement employee and of which two DDC employees would be members; and (iii) to appoint a HeidelbergCement employee as project manager for the Transaction.

(68) Ninth, HeidelbergCement submitted the indicative offer for the purchase of Cemex Austria, Hungary and Croatia on 8 June 2015 after receiving oral approval from Schwenk. In that offer, HeidelbergCement referred to itself as [content of internal document].

(69) Tenth, the indicative offer highlighted HeidelbergCement's strategic interest in the Transaction: [content of internal document].

(70) Eleventh, the indicative offer highlighted how the Transaction is in line with HeidelbergCement's broader corporate strategy: [content of internal document]. It is particularly striking in this respect that the Transaction was considered to strengthen the [content of internal document] of HeidelbergCement. This contradicts HeidelbergCement's argument that it had no "additional or different interest in the transaction than DDC" and that "HeidelbergCement's […] interest was nothing more than […] the interest of DDC".

(71) Twelfth, a presentation prepared by [name of HeidelbergCement employee] in July 2015 further confirms that HeidelbergCement perceives the Transaction as serving its own interest: [content of internal document].

3.4.2. Organisation of the Transaction

(72) HeidelbergCement organised the Transaction, including developing the business case and the transaction structure, preparing the deal valuation and leading the final negotiations with Cemex. Schwenk was kept informed regularly about the organisation of the Transaction by HeidelbergCement and never sought to oppose HeidelbergCement's role in any way. DDC strictly adhered to decisions taken by HeidelbergCement. That is confirmed by a number of internal documents submitted by the Notifying Parties and DDC.

(73) First, sometime before the meeting of [names of HeidelbergCement employees] on 6 May 2015, HeidelbergCement and Schwenk decided to pursue the acquisition of Cemex Hungary and Cemex Croatia with DDC as purchaser.

(74) Second, sometime before the meeting of [names of HeidelbergCement employees] on 6 May 2015, HeidelbergCement took decisions regarding the implementation and composition of the steering committee, the submission, preparation and timing of an indicative offer, the structure of the due diligence and the related responsibilities.

(75) Third, the HeidelbergCement members of the steering committee attended negotiations with Cemex and prepared detailed documentation, deal valuation and other components of the business case for the decision by the HeidelbergCement management board and supervisory board to approve the acquisition.

(76) Fourth, HeidelbergCement negotiated the NDAs with Cemex, organised and conducted the due diligence and organised the implementation planning.

(77) Fifth, prior to the signature of the written SPA on 11 August 2015, a verbal agreement on the main terms of the Transaction was reached between [name] (HeidelbergCement) and [name] (Cemex) on 31 July 2015.

(78) Sixth, following the verbal agreement on the main terms of the Transaction, [name of HeidelbergCement employee] negotiated certain open issues directly with [name of Cemex employee]: [content of internal document]. "At that time, the mission of the 2015 steering committee (manage offer procedure, coordinate due diligence, negotiate contracts) was completed except for the merger control clearances."

(79) Seventh, since June 2016, HeidelbergCement and Schwenk representatives are members of a steering committee for the integration of the Transaction whose role was to prepare [content of internal document] in relation to [content of internal document].

(80) Eighth, Schwenk sought and received updates about the progress of the Transaction on a weekly basis.

(81) Ninth, in June 2015, Schwenk discussed with HeidelbergCement the potential structure of the Transaction.

3.4.3. Financing of the Transaction

(82) HeidelbergCement and Schwenk designed the financing and related corporate structure of the Transaction. Furthermore, HeidelbergCement selected the banks that should be contacted and took decisions on the allocation of debt levels. This is confirmed by a number of internal documents submitted by the Notifying Parties and DDC.

(83) First, it was decided at a meeting on 6 May 2015 attended exclusively by HeidelbergCement employees that one of the two HeidelbergCement representatives on the steering committee would be responsible for keeping "the contact with Cemex and the banks".

(84) Second, in May 2015, Heidelberg envisaged scenarios for the financing of the Transaction where HeidelbergCement and Schwenk would pay EUR […] million out of a purchase price that was estimated at that time to be EUR […] million. The contribution of HeidelbergCement and Schwenk was, however, subsequently reduced to EUR […] million each, in which case an "EU filing" would not be required provided that the [content of internal document].

(85) Third, in May 2015, HeidelbergCement initiated contacts with consultancy firms for the financial due diligence.

(86) Fourth, at the end of July 2015, [content of an internal document on HeidelbergCement’s financial commitment to the Transaction] for Cemex Croatia and Cemex Hungary.

ID2624, reply to request for information of 19 January 2017.

ID1280, Parties' reply to RFI 28 October 2016, Q46b and Q46c.

ID1103-533, e-mail of […] to […] of 17 June 2015.

ID1103-2202, e-mail of […] to […] of 26 June 2016: [content of internal documents].

ID437-93, [internal document].

ID1088-5876, Submission of information - GL_C001_00052126, slides 2, 4 and 5: under the first scenario, HeidelbergCement and Schwenk would contribute EUR […] million each, under the second scenario EUR […] million each; slide 2 of that document: the transaction value was considered to be EUR […] million (EUR […] million for Cemex Hungary and EUR […] million for Cemex Croatia); ID437-94, M.7878_Form CO_Annex 5.4.-I.-(02)_Meeting_Minutes_ Kickoff.PDF, point 13: [content of internal documents], as the transaction was valued at EUR […] million (see point 11), the remainder would be EUR […] million.

ID437-99, M.7878_Form CO_Annex 5.4.-I.-(07)_Presentation_Cemex_Assets_in_Southeast_Europe.PDF.

ID1103-4750, Response to questionnaire 2 of 28 October 2016. - M7878_All_files - GL_C001_00049950, Letter […] of 15 May 2015; ID225-10728, GL_C001_00024933.pdf, letter dated June 2015 of KPMG to "HeidelbergCement AG".

(87) Fifth, in August 2015, Schwenk indicated its willingness to grant [business secret] to avoid that HeidelbergCement and Schwenk would need to issue guarantees towards the banks to secure the financing by DDC.

(88) Sixth, HeidelbergCement decided which entity should take loans, whether a new entity ("NewCO") should be established for these purposes, which company should be the direct acquirer, which companies' capital should be increased and whether HeidelbergCement through its subsidiary HC CEE which is the direct shareholder on HeidelbergCement's side in DDC would need to inject more funding:

-"[Content of internal document]."

-"[Content of internal document]."

-"[Content of internal document]."

-"[Content of internal document]."

(89) Seventh, HeidelbergCement engaged with banks in July 2015, prepared an information memorandum for the banks, negotiated the NDAs with the banks between July and October 2015 and attended meetings with the banks in November 2015.

(90) Eighth, on 31 July 2015, HeidelbergCement agreed with Cemex on the final purchase price.

(91) Ninth, in August 2015, HeidelbergCement required DDC to secure the agreed financing according to HeidelbergCement's guidelines with close involvement of HeidelbergCement's CFO and/or HeidelbergCement's Group Treasurer: [content of internal document]. Concerning HeidelbergCement's hesitation to grant corporate guarantees, HeidelbergCement explained to DDC in this context: [content of internal document].

ID437-101, M.7878_Form CO_Annex 5.4.-I.-(09)_Supervisory_Board_Presentation_Erwerb_Cemex-Aktivitaeten_Suedosteuropa.PDF: [content of internal document] (curtesy translation by the Commission: [content of internal document]).

ID1088-2015, e-mail of […] to […] of 10 August 2015 [content of internal documents].

ID1103-5865, Response to questionnaire 2 of 28 October 2016. - M7878_All_files - GL_C001_00052111. See for the organisation structure: ID1088-3255, GL_C001_00046164.pptx, slides 6 and 7.

Ibid.

Ibid.

ID1103-836, Response to questionnaire 2 of 28 October 2016. - M7878_All_files - GL_C001_00023204.

ID225-12210, M.7878 - QP 2 Section B (E-mails request) - USB Stick - GL_C001_00028451, […], e-mail of 30 July 2015: [content of internal documents].

(92) Tenth, at the end of September 2015, decisions about a debt push-down strategy were taken by a project team consisting mainly of HeidelbergCement representatives, namely [name] (Controlling CECA), [name] (Group Treasury HeidelbergCement) and [name] (Tax Manager HeidelbergCement).

(93) Eleventh, HeidelbergCement was, and still is, willing to provide "soft comfort letters" to the financing banks stating "that they will not sell shares of DDC to third parties as long as DDC is still obligated under the loan agreement, and otherwise they will use (only) its best efforts to cause DDC to repay the outstanding amounts before any such sale."

(94) Twelfth, Schwenk was, and still is, willing to provide "soft comfort letters" to DDC "to support DDC’s negotiations with the banks by showing that SCHWENK is one of DDC’s shareholders. These comfort letters contain no funding obligation or other financial contribution by SCHWENK".

3.4.4. Schwenk's involvement in the Transaction

(95) Contrary to what the Notifying Parties claim, Schwenk's involvement in the initiation, organisation and the financing of the Transaction was significant and not limited to the role of a shareholder exercising its mandatory rights in a joint-venture.

(96) First, at some point before 6 May 2015, HeidelbergCement contacted Schwenk to ensure that it was "in favour of the deal".

(97) Second, in May 2015, Heidelberg was envisaging scenarios for the financing of the Transaction where HeidelbergCement and Schwenk would pay EUR […] million out of a purchase price that was estimated at that time to be EUR […]. The contribution of HeidelbergCement and Schwenk was, however, subsequently reduced to EUR […] million each, in which case an "EU filing" would not be required provided that the [content of internal document].

(98) Third, in June 2015, Schwenk discussed with HeidelbergCement about the potential structure of the Transaction.

(99) Fourth, on 8 June 2015, HeidelbergCement submitted the indicative offer for the purchase of Cemex Austria, Hungary and Croatia after receiving oral approval from Schwenk.

ID1103-2891, Response to questionnaire 2 of 28 October 2016. - M7878_All_files - GL_C001_00045228.

ID1088-6802, "Group Tax – internal Letter" of 28 September 2015, prepared by […].

ID2600,RE_ M.7878 - HEIDELBERGCEMENT _ SCHWENK _ CEMEX HUNGARY _ CEMEX CROATIA - RFI Financing (HC_DDC).pdf, reply to question 2.

ID2594, Reply : RFI Financing (Schwenk) - AW_ M.7878 - HEIDELBERGCEMENT _ SCHWENK _ CEMEX HUNGARY _ CEMEX CROATIA - RFI Financing (Schwenk), reply to question 1.

(100) Fifth, in August 2015, Schwenk indicated its willingness to grant a [business secret] to avoid that HeidelbergCement and Schwenk would need to issue guarantees towards the banks to secure the financing by DDC.

(101) Sixth, Schwenk sought and received updates about the progress of the Transaction on a weekly basis.

(102) Seventh, Schwenk was, and still is, willing to provide "soft comfort letters" to DDC [content of internal document].

(103) Eighth, Schwenk was involved not only in matters of general strategic importance for the Transaction but also in details of the implementation of the Transaction, including membership of a steering committee for the integration of the Transaction.

(104) Ninth, it is legally irrelevant whether Schwenk may have been involved to a different degree than HeidelbergCement in the Transaction. In the same way that two undertakings may be classified as a leader because they have been a "significant driving force" in a cartel, two parents of a joint venture may have a significant, albeit different, involvement in a concentration.

(105) Tenth, in a situation where one of the two parent companies of a full function joint-venture is more significantly involved in the initiation, organisation and financing of a concentration, it is sufficient, in order for the involvement of the second parent company to be significant, that it knows of the involvement of the first parent company and does not oppose that involvement. In any event, in this case, Schwenk's involvement went further than limiting itself to not opposing HeidelbergCement's more significant involvement (see recitals (96)-(103)).

3.4.5. DDC's involvement in the Transaction

(106) Contrary to what the Parties claim in their replies to the Article 6(1)(c) Decision, the Statement of Objections and the Letter of Facts, DDC neither initiated nor organised the Transaction.

(107) First, whether DDC and its subsidiary Tvornica cementa Kakanj d.d. ("TCK") had considered expanding their business via the acquisition of Dalmacijacement in 2003 or Nexe in 2013 is irrelevant to the assessment of the undertakings that, in 2015, initiated, organised and financed the Transaction.

(108) Second, while HeidelbergCement and Rohrdorfer were represented in the SPA negotiation team through members of their management, no DDC management member was part of the negotiation team; DDC was merely represented through its external legal counsel.

ID1088-2015, e-mail of […] to […] of 10 August 2015 [content of internal documents].

ID1103-533, e-mail of […] to […] of 17 June 2015.

ID2594, Reply : RFI Financing (Schwenk) - AW_ M.7878 - HEIDELBERGCEMENT _ SCHWENK _ CEMEX HUNGARY _ CEMEX CROATIA - RFI Financing (Schwenk), reply to question 1.

ID1280, Parties' reply to RFI 28 October 2016, Q46a and Q46c.

(109) Third, contrary to what was initially argued by HeidelbergCement during the administrative procedure, the e-mail of 7 May 2015 sent by [name] of DDC to [name] of HeidelbergCement contains no indication that "[i]t was [name of DDC board member] (Chairman of TCK) who initiated the project and who undertook to inform and convince [name of HC board member] accordingly by sending him a presentation from 2003 and 2004 when he initiated a discussion if TCK should acquire what are nowadays Cemex assets in Croatia".

(110) On the contrary, the wording of the email indicates that DDC was simply providing "[content of internal documents]" to HeidelbergCement as to how to best "[content of internal documents]": "[content of internal documents]". HeidelbergCement subsequently confirmed during the administrative procedure that it was [name of HeidelbergCement employee] that informed [name of DDC employee] about the sale of Cemex Croatia and not [name of DDC employee] that informed [name of HeidelbergCement employee].

(111) Moreover, even if [name of HeidelbergCement employee] may have consulted [name of DDC employee] on whether the "chances of such investment should be further explored and if [DDC] still think[s] this would be interested [sic] acquisition" this demonstrates neither the significance of DDC's involvement in the initiation of the Transaction, nor the lack of significance of the HeidelbergCement's and Schwenk's involvement.

(112) Fourth, the fact that DDC "does not have a legal department and lacks manpower and experience required to negotiate a large M&A transaction", cannot explain the significant involvement in the Transaction by HeidelbergCement and Schwenk. Revealingly, the Notifying Parties have provided no documentary evidence either of DDC requesting help from HeidelbergCement and Schwenk to pursue the Transaction or that it was for DDC to take the final decision regarding the organisation of the Transaction and that thus DDC ultimately remained "in control of the transaction" while HeidelbergCement had a mere advisory role. On the contrary, DDC's role was of a supportive character, fulfilling tasks that were allocated to it. This functional support by DDC does not demonstrate its significant involvement in the initiation and organisation of the Transaction.

(113) Fifth, DDC was not seen as an independent acquirer but as a mere "vehicle" for the Transaction: "[content of internal document]."

(114) Sixth, decisions about a debt push-down strategy were taken by HeidelbergCement and not DDC (see recital (92)).

ID1280, Parties' reply to RFI 28 October 2016, Q47.

ID1280, Parties' reply to RFI 28 October 2016, Q47 and ID1103-599.

ID2668, e-mails of […] dated 24 January 2017; ID3696, e-mail of […] dated 8 March 2017; ID3697, Witness Statement_[name].pdf.

ID3697, Witness Statement_[name].pdf.Witness Statement_[name of DDC board member].pdf.

ID2415, M7878_HC DDC_Comments on SO_FINAL_20170103.PDF, paragraph 27.

ID2901, HeidelbergCement's reply to the Letter of Facts. The only evidence submitted in this regard is a "Written Witness Statement" of [name] dated 30 January 2017 and submitted on 8 March 2017, ID3697, Witness Statement_[name].pdf. In the absence, however, of any contemporaneous documentary evidence from the time of the initiation phase of the Transaction, the probative value of the witness statement is limited.

ID1088-4344, 1st Q&A GCF of 26 June 2015.

(115) Seventh, the claim that "no decision regarding the structuring or financing of the deal was or would have been taken without the agreement of [name of DDC board member] and the other representatives of DDC" is unsubstantiated.

3.4.6. Conclusion

(116) It follows, that the Transaction has a Union dimension within the meaning of Article 1 of the Merger Regulation since the undertakings concerned have a combined aggregate world-wide turnover of more than EUR 5 000 million (HeidelbergCement: EUR […] in 2015; Schwenk EUR […] in 2015; Cemex Hungary: EUR […] in 2015; Cemex Croatia: EUR […] in 2015). Two of them have a Union-wide turnover in excess of EUR 250 million (HeidelbergCement: EUR […] in 2015; Schwenk: EUR […] in 2015), but they do not achieve more than two thirds of their aggregate Union-wide turnover within one and the same Member State.

(117) By reasoned submission of 25 May 2016, the Notifying Parties requested a partial referral pursuant to Article 4(4) of the Merger Regulation. The Notifying Parties requested the Transaction to be examined by the competent authorities of Hungary as regards its effects on Hungarian markets. On 22 June 2016, the Commission referred the assessment of the effects on the relevant markets in Hungary to be examined by the Hungarian Competition Authority, pursuant to Article 4(4) of the Merger Regulation. Therefore, only the remainder of the Transaction will be assessed by the Commission.

(118) Prior to the notification of the Transaction, the Commission sent requests for information to the Parties and their largest competitors selling grey cement in Croatia. In response, the Parties and third parties submitted detailed information on their relevant cement plants, grinding stations and terminals in terms of reserves, capacity, production, sales, production costs, prices, revenues and transport costs. In addition, the Commission requested the Notifying Parties to submit an extensive selection of internal documents.

(119) Following notification of the Transaction and during the initial investigation, the Commission sent more than 170 written requests for information to the Parties' customers as well as to the Parties' competitors in Croatia. The Commission also requested further information from the Notifying Parties, including additional internal documents.

(120) During its in-depth investigation, the Commission continued its analysis of competitors' data and requested further information from nearly all of the Parties' competitors in Croatia. Furthermore, the Commission addressed written requests for information to additional customers – including the Parties' most important competitors in the sale of ready-mix concrete in Dalmatia – and held phone calls with a number of customers and competitors. The Commission also contacted port operators on the Croatian coast. The Commission also received several submissions from, and held several meetings with, the Parties.

ID3697, Witness Statement_[name].pdf.

(121) Following the offer of the commitments on 26 January 2017, the Commission addressed requests for information to more than 170 customers and competitors as well as port operators.

(122) The Commission also had contacts with the Hungarian, Croatian and Bosnian competition authorities.

(123) For the reasons set out in this Decision, the Commission concludes that the Transaction will significantly impede effective competition through non-coordinated effects, which could amount in particular to the creation of a dominant position, in grey cement markets alternatively defined as (i) a 250km circular catchment area around Cemex's Split plant or (ii) a modified 250km catchment area around Cemex's Split plant. Both of those markets include the overlap lens with the 250km circular catchment area of the Kakanj plant.

(124) Consequently, after providing an overview of the Parties' activities in Croatia in Section 5, the Commission will define the relevant product and geographic markets with respect to grey cement in Section 6 before setting out in Sections 7 and 8 why the Transaction will significantly impede effective competition in a substantial part of the internal market. In Section 9, the Commission will then explain why the commitments submitted by the Notifying Parties do not render the Transaction compatible with the internal market. Finally, Section 10 concludes.

5. THE PARTIES' ACTIVITIES IN CROATIA

(125) The Parties mainly sell grey cement and ready-mix concrete in Croatia.

(126) Cemex Croatia operates three cement plants in south-eastern Croatia near Split, all located within a 10km area. Cemex Croatia [strategic information and capacity figures]. Cemex Croatia also operates two rail depots in Zagreb (Maksimir and Podsued), one sea terminal in north-western Croatia (Bakar) and one sea terminal in the south of Croatia (Metković). Cemex Croatia exports [strategic information] of its production to [strategic information].

(127) HeidelbergCement and DDC compete in Croatia through cement imports. DDC supplies the North-East of Croatia from its plant in Beremend (Hungary), and the South-East from its plant in Kakanj (Bosnia-Herzegovina). In the North-West, Italcementi, now owned by HeidelbergCement, imports grey cement from its grinding station in Trieste (Italy).

(128) Both DDC and Cemex Croatia are to a limited extent vertically integrated into the production of ready-mix concrete in Croatia, with, respectively, 6 and 5 ready-mix concrete sites.

(129) DDC's and Cemex' grey cement production plants are set out on the map in Figure 1. The customer locations of Cemex Split and DDC Kakanj – which are most relevant for this Decision – are set out in Figure 2.

6. RELEVANT MARKETS

(130) The remainder of the Transaction, reviewable by the Commission, mainly concerns the markets for grey cement in Croatia.

6.1. Description of the products as well as production and distribution assets

(131) Cement is one of the main input products in modern construction. It is produced by grinding clinker and alternative cementitious materials. Clinker is the main ingredient in the production of cement and is produced at high temperatures from limestone and other constituents in cement kilns. In some cases, mineral components and other cementitious materials are added to the mixture by either grinding them together with clinker or blending separate ground materials together.

Based on the map contained in ID437-255 M.7878_Form CO_Annex 6.III.1.-Grey Cement_HR_[advisor's identity] economic analysis_revised.PDF, page 11.

ID441-169, Annex 5.4.III.-RFI 20160608_QP1_QA.01_[advisor's identity] memo on geographic market definition, page 6.

The descriptions in this section are taken from the decision in Case M.7744 – HeidelbergCement/Italcementi, recitals 9-20 and the decisions referenced in those recitals.

In this Decision, the term alternative cementitious materials refers to substances other than clinker that have cementitious or hydraulic binding properties and that are used as supplementary materials in the production of cement and concrete. The most common alternative cementitious materials are fly ash (a by-product of coal combustion in thermal power plants) and blast furnace slag (a by-product in the production of iron) in the form of granulated blast furnace slag ("GBS") or ground granulated blast furnace slag ("GGBS"). They are added to cement and concrete in order to impart specific characteristics to the end product and to substitute, on the one hand, clinker in the production of cement and, on the other hand, cement in the production of concrete.

(132) Cement is used in the building and construction sector mainly to bind other materials for stability and strength as well as an intermediary product mostly for the production of ready-mix concrete, concrete products and mortar.

(133) There are two main types of cement: white cement and grey cement. The main difference between white and grey cement lies in the particular quality of limestone used for the production of white cement. Furthermore, white cement is used for different purposes (in particular reflecting esthetical/optical aspects). White cement is produced in comparably limited quantities and is more expensive than grey cement.

(134) Within grey cement, there are different classes available and further grades can be produced according to customer requirements. Cement classes are defined by strength development and setting times, which are in turn determined by the proportions and nature of the different raw cementitious products used to make that particular cement type. The Union standard EN 197-1 defines five classes of common cement that comprise Portland cement as a main constituent.

Table 1 - Classes of common cement according to EN 197-1

CEM I Portland cement comprising Portland cement and up to 5% of minor additional constituents

CEM II Portland-composite cement Portland cement and up to 35% of other single constituents

CEM III Blast furnace cement Portland cement and higher percentages of blast furnace slag

CEM IV Pozzolanic cement Portland cement and up to 55% of pozzolanic constituents (volcanic ash)

CEM V Composite cement Portland cement, blast furnace slag or fly ash and pozzolana

(135) Cement is sold both in bulk and in bags. Bags containing about 25-30 kg of cement are sold through do-it-yourself stores and building material retailers whereas bulk cement meets the demand of ready-mix plants, plants producing concrete products and building sites.

(136) There are generally three types of cement production sites: integrated cement plants, grinding stations and terminals (also referred to as blending stations).

(137) An integrated cement plant is a manufacturing facility that covers the entire cement production process from the mining of raw materials to the dispatching of cement. The production process involves the following steps: (i) raw material extraction or mining from a quarry; (ii) raw material preparation and blending; (iii) raw feed preparation out of the raw materials in the form of meal; (iv) clinker production, which forms the main process of an integrated plant, that is to say converting raw feed in a thermochemical reaction in a cement kiln into the desired calcined mineral ("clinker"); (v) grinding and blending of clinker with gypsum or other components (such as alternative cementitious materials) into the desired cement product; and (vi) storage and handling of cement products, including dispatch.

ID1993, UK Competition Commission, "Anglo American PLC and Lafarge S.A. A report on the anticipated construction materials joint venture between Anglo American PLC and Lafarge S.A.", 1 May 2012.

Commission communication in the framework of the implementation of Regulation (EU) No 305/2011 of the European Parliament and of the Council of 9 March 2011 laying down harmonised conditions for the marketing of construction products and repealing Council Directive 89/106/EEC (Publication of titles and references of harmonised standards under Union harmonisation legislation), OJ C 259, 8 August 2014, page 1; see also http://standards.cen.eu/dyn/www/f?p=204:110:0::::FSP PROJECT:27250&cs=13B3EDD735E572AF56B7EC3A4CA2E1AF6.

(138) A grinding station or grinding mill does not include the mining and the thermal process of producing clinker, but only the final grinding, blending and handling steps, with clinker and other raw materials being delivered from a separate plant or sourced elsewhere.

(139) A terminal is a depot facility which is typically accessible by navigable water, railway or by road. A terminal consists of a relevant transport platform and of a silo-type storage installation and is a strategic asset enabling a cement manufacturer to supply a territory where it does not operate a production site.

6.2. Product market definitions

6.2.1. Previous Commission decisions

(140) In past decisions, the Commission has defined distinct product markets for grey cement, aggregates and ready-mix concrete. The Commission has also defined distinct product markets for white cement and grey cement.

(141) Concerning grey cement, the Commission has not further segmented the market according to grades or classes (CEM I to CEM V), mainly since such distinction did not have an impact on the outcome of the competitive assessment. Moreover, the Commission also left open whether the market for grey cement could be further segmented according to whether grey cement is sold in bulk or bagged.

6.2.2. The Notifying Parties' submissions

(142) While there are distinct product markets for white cement and grey cement, the Notifying Parties consider that the distinction between bagged and bulk cement can be left open and that in any event all major cement producers supply cement in both bagged and bulk form.

(143) Furthermore, the Notifying Parties submit that while generally the market for grey cement should not be further segmented according to grades or classes (CEM I to CEM V), grades and classes may still play a role in the analysis of specific market situations.

6.2.3. The Commission's assessment

(144) For the reasons set out in recitals (146) to (159), the Commission concludes that while there are distinct product markets for white cement and grey cement, the exact sub-segmentation of the grey cement market (bagged versus bulk cement, between different cement types and grades) can be left open since the Transaction will lead to a significant impediment of effective competition under all plausible product market definitions.

Cases M.7550 – CRH/Holcim Lafarge divestment business, 24 April 2015, recitals 35-36; M.7252 – Holcim/Lafarge 15 December 2014, recital 10; M.7054 – Cemex/Holcim Assets, 9 September 2014, recital 39; M.7009 – Holcim/Cemex West, 5 June 2014, recitals 24 and 41; M.3713 – Holcim/Aggregate Industries, 14 March 2005, recital 7; M.3572 – Cemex/RMC, 8 December 2004, recital 11; M.3415 – CRH/SEMAPA/Secil JV, 28 May 2004, recital 10; M.2317 – Lafarge/Blue Circle (II), 1 March 2001, recital 9; M.1157 – Skanska/Scancem, 11 November 1998, recital 31.

Cases M.7009 – Holcim/Cemex West, 5 June 2014, recital 14, M.7252 – Holcim/Lafarge, 15 December 2014, recital 36.

Cases M.7252 – Holcim/Lafarge, 15 December 2014; recitals 53-56; M.7054 – Cemex/Holcim Assets, 9 September 2014, recital 43; M.7009 – Holcim/Cemex West, 5 June 2014, recital 49.

(145) The competitive assessment, however, takes into account: (i) the fact that bulk cement represents 70% of sales in Croatia; (ii) the fact that some suppliers are able only to supply bagged cement for logistical reasons; (iii) the differentiation of suppliers in terms of cement classes; and (iv) the particular relevance of cement type CEM II in Croatia.

6.2.3.1. Bagged and bulk cement

(146) In Croatia currently 70 % of grey cement is sold in bulk and 30% of grey cement is sold in bags. [Strategic information].

(147) On the one hand, there are a number of factors suggesting that there is no need to segment further the market for grey cement between bulk and bagged cement.

(148) First, the physical properties of the final product are the same and it is only the mode of delivery that differs.

(149) Second, no supplier of cement in the Broader Region specializes in the production of only bagged or bulk grey cement since they have the necessary bagging and palletising installation in place.

(150) On the other hand, there are a number of factors suggesting that it could be necessary to segment further the market for grey cement between bulk and bagged cement.

(151) First, the price of bagged cement can be significantly higher, in particular due to additional packaging costs.

(152) Second, the costs for transporting bulk cement can be significantly higher:

According to LafargeHolcim: "The difference in transport costs for bagged vs bulk cement arises from lower tariffs charged per km for bagged cement transport as compared to bulk cement transport. This arises because the expectation is that the delivery trucks used for bagged cement are smaller and can, after delivering the cement, find backhaul routes easier thereby optimising their cost. For bulk cement transport, by contrast, the size of the trucks required to deliver makes transport backhauling almost non-existent."

According to the Parties: "Bulk cement is transported in special vehicles (truck + silo), which have a higher payload in terms of traffic law limitations and allow loading up to 27 tonnes. This may lead to a lower unit transport price per tonne for bulk cement than for bagged cement. On the other hand, cement silo trucks are specialised vehicles only used for cement transport. As a

ID1105-1, presentation 'Sales & Logistics – Overview, September 2016'. The split between bagged and bulk cement is similar when considering the 250km circular catchment area around the Cemex' plant in Split ([strategic information]) and the midpoint between the Parties' cement plants in Split and Kakanj ([strategic information]).

DDC and Cemex sell in Croatia [sales volume]%, in Dalmatia [sales volume]% and in Slavonia [sales volume]% of their cement in bulk.

consequence, the number of available trucks is limited. From this follows also that after unloading cement at the customer location, the truck must return without loading or requires special cleaning which might increase transport costs (especially over longer distances).

Bag cement can be transported in regular trucks. Those may carry up to 24 tonnes of palletized bagged cement which may lead to higher unit transport costs per tonne compared to bulk cement. On the other hand, the number of such regular truck available is much higher than the number of special trucks used for bulk cement. Furthermore, there is the possibility to combine cement deliveries with return transport of other commodities transports, which in turn may decrease transport costs.

(153) Third, nearly all the customers that responded to the market investigation indicated that they are unable to switch from bagged to bulk cement (and vice versa).

(154) Fourth, customers of bulk cement may prefer closer suppliers as compared to bag cement customers in order to provide security of supplies.

(155) Fifth, the identity and market shares of the competitors differ to a certain extent between the supply of bulk cement and bagged cement.

6.2.3.2. Different classes of cement

(156) Cement type CEM II is with approximately 89% by far the most important cement type supplied in Croatia. It accounts for approximately […]% of Cemex Croatia's overall cement sales and for approximately […]% of DDC's cement sales.

(157) There are arguments both for and against a possible further segmentation of the market for grey cement between different cement classes.

(158) On the one hand, from the demand side, most competitors indicated that their customers can switch between different classes of grey cement, although there are differences per customer groups with some specialised customers (pre-cast concrete, mortar) reluctant to change cement classes as that would entail changes in their production recipes.

(159) On the other hand, from the supply-side, most producers of grey cement indicated that they can switch their production to another cement class, at least for certain classes. In general, there are no sophisticated specialisations for different types of cement that would require different production systems to be in place.

ID1148, HC/DDC's and Cemex' combined response to the RFI to the Parties dated 10 November 2016, Q7.

See response to Questionnaire to Customers (Q1), question 7.

ID1833, reply of Titan to request for information of 09 November 2016, Q23.

6.3. Geographic market definition

(160) For the reasons set out in recitals (163) to (206), the Commission concludes that the definition of the relevant geographic market can be left open as the Transaction will significantly impede effective competition whether the relevant market is defined as either of the following:

(1) circular catchment areas of 250km around the Parties' plants; or

(2) modified 250km catchment areas around the Parties' plants.

(161) The Commission also concludes that it can be left open whether those two alternative market definitions should include non-EEA territory (in particular Bosnia-Herzegovina) as in any event, the competitive assessment focusses only on the parts of the relevant markets in the EEA.

(162) The competitive assessment, however, takes into account the fact that, while the areas covered by those two alternative market definitions are sufficiently homogenous and can be distinguished from neighbouring areas, there are variations in competitive conditions within each of those two alternative markets.

6.3.1. Previous Commission decisions

(163) In previous decisions, the Commission has considered that the geographic market for grey cement consists of a group of geographic markets centred on different cement plants, overlapping one another.

(164) The scope of the relevant geographic markets was determined by the distance from the plant at which cement may be sold. Generally, radii of 150 km and/or 250 km were taken into account based on a rough assessment of how far 70%/80%/90% of grey cement is shipped based on sales data. When relevant, the Commission also assessed whether the circular catchment areas on the basis of such 150km or 250km radii should include territories across national borders. The established radii were also applied to competing suppliers' plants which allowed identifying those competitors' plants whose circular catchment areas overlapped with the relevant markets and thus identifying in a systematic way the immediate competitive constraints facing the merged entity.

Cases M.7744 – HeidelbergCement/Italcementi, recitals 39-41; M.7550 – CRH/Holcim Lafarge divestment business, 24 April 2015, recitals 47-48; M.7252 – Holcim/Lafarge, recitals 59-63; M.7054 – Cemex/Holcim Assets, 9 September 2014, recital 52; M.7009 – Holcim/Cemex West, 5 June 2014, recital 64; M.460 – Holdercim/Cedest, 6 July 1994, recital 16; M.1030 – Lafarge/Redland, 16 December 1997, recital 16; M.1157 – Skanska/Scancem, 11 November 1998, recital 56; M.2317 – Lafarge/Blue Circle (II), 1 March 2001, recital 8; and M.3572 – Cemex/RMC, 8 December 2004, recital 20.

Cases M.7009 – Holcim/Cemex West, 5 June 2014, recital 14, M.7252 – Holcim/Lafarge, 15 December 2014, recital 36.

Cases M.7252 – Holcim/Lafarge, 15 December 2014; recitals 53-56; M.7054 – Cemex/Holcim Assets, 9 September 2014, recital 43; M.7009 – Holcim/Cemex West, 5 June 2014, recital 49.

(165) Based on those circular catchment areas, the Commission also drew circular catchment areas around the mid-points between plants. Furthermore, the Commission identified the specific overlap areas and, in some cases, focused its assessment in particular on such overlap areas which represented only parts of the circular catchment areas drawn around plants. The purpose of those additional steps was to identify the areas in which plants compete or could compete and to eliminate from the assessment areas where only one plant does or can compete.

(166) The Commission's case practice also acknowledged that additional corrections to the circular catchment areas may have to be made according to the actual supply patterns which can be influenced by topography and other factors.

6.3.2. The Notifying Parties' submissions

(167) The Notifying Parties submit that the relevant geographic markets in this case should be defined as circular areas of 250km around the relevant grey cement plants, reflecting the distance up to which suppliers can economically sell grey cement in Croatia and the Broader Region. The Notifying Parties put forward a number of reasons.

(168) First, the distribution of third party grey cement sales of Cemex Croatia around Split would support using a 250km radius as 70% of sales are made in a radius of […]km and 90% in a radius of […]km (both without seaborne transport). This is due [strategic information].

(169) Second, the Commission should consider not only selected production sites in the wider region but take into account only Croatian plants in its analysis of the delivery distances.

(170) Third, despite cross-border sales from Croatia to Bosnia-Herzegovina and Serbia, non-EEA territory from the 250km circular catchment areas should be excluded from the competitive assessment.

(171) Fourth, Croatia should not be divided into regional markets (Istria and Kvarner, Dalmatia, Central Croatia and Slavonia). Such regional internal subdivisions, as referred to in the Parties' internal documents, are based on historical and administrative reasons and do not reflect commercial reality.

6.3.3. The Commission's assessment

(172) The Commission concludes that it can be left open whether the relevant market should be defined as either of the following:

(1) circular catchment areas of 250km around the Parties' plants; or

(2) modified 250km catchment areas around the Parties' plants.

M.7744 – HeidelbergCement/Italcementi, recital 91; M.7252 – Holcim/Lafarge, recitals 102-109.

Case M.7744 – HeidelbergCement/Italcementi, recital 40; see also cases M.7252 – Holcim/Lafarge, recitals 61-63; M.7550 – CRH/Holcim Lafarge divestment business, recital 48, M.7054 – Cemex/Holcim Assets, recitals 57-58; M.7009 – Holcim/Cemex West, recitals 69-70.

(173) After setting out the general principles for assessing geographically differentiated markets such as grey cement makers (section 6.3.3.1), this section: (i) explains why the appropriate radius for the circular catchment areas is 250km (section 6.3.3.2); (ii) assesses the role of national borders to determine whether territories inside and outside Croatia fall within the relevant catchment areas (section 6.3.3.3); and (iii) conducts a more detailed analysis of supply and demand conditions by analysing the catchment areas on the basis of actual road delivery distances and delivery patterns (section 6.3.3.4).

6.3.3.1. General principles for assessing geographically differentiated markets such as grey cement markets

(174) Grey cement is a heavy and bulky but rather low-value product. As a result, the proportion of transport costs in total variable costs of supplying cement to a customer can be significant and that proportion increases with the distance to the customer. That impact is reinforced by security of supply considerations that make delivery more difficult with increasing distance. Those factors limit the distance over which grey cement can be economically shipped.

(175) Accordingly, competitive conditions will change gradually for customers in different locations. With any move on the map, the competitive constraint imposed by the surrounding cement suppliers will change in function of the change in travel distance (and hence transport costs and security of supply) from each supplier to the customer. Moreover, prices for grey cement are individually negotiated with customers allowing suppliers to charge different prices to different customers.

(176) Those conditions mean that while competitive conditions within a certain circular catchment area may be sufficiently homogenous and can be distinguished from neighbouring areas, competitive conditions may be more homogeneous between customers located close to one another and less homogeneous between customers located further away.

(177) In order to reflect those differences, the Commission applies the following different methods (often alternatively at the same time):

– drawing circular catchment areas around production plants and identifying the overlap lens between those circles;

Figure 3: Identifying the overlap lens of catchment areas

ID444, Form CO, paragraph 198.

– drawing circular catchment areas around mid-points between production plants (that is the mid-point of a straight line drawn between the two plants);

– drawing circular catchment areas around customers, followed by an aggregation (such as for the customers in the overlap lens [the striped area in Figure 3] or for customers in certain clusters) to compute local measures of concentration.

6.3.3.2. Appropriate radius for the circular catchment areas

(178) For the reasons set out in recitals (179) to (183), the Commission concludes that the appropriate radius for the circular catchment areas around the Parties' plants should be 250km.

(179) First, the Commission has based itself on the data of the Parties and other suppliers regarding delivery distances by rail and road in Croatia. Only LafargeHolcim currently uses seaborne deliveries to reach customers in Croatia. Furthermore, and as will be set out in section 7.4.2, seaborne deliveries in Croatia are currently not an alternative for suppliers other than Cemex and LafargeHolcim due to the lack of appropriate infrastructure in Croatian ports.

(180) Second, as set out in Table 2, 70% of the Parties' sales are delivered on average at a geodesic distance of up to […]km and 90% of sales are delivered on average at a geodesic distance of up to […]km. Based on road distances, 70% of sales are delivered at up to […]km on average, and 90% of sales are delivered at up to […]km on average.

Table 2 - Range, median and average of the delivery distances of Parties

70% of sales – 90% of sales – 70% of sales – geodesic distance geodesic distance road distance

90% of sales – road distance

Range [50-100]-[150-200]km

[100-150]km-[350-400]km

[100-150]km-[250-300]km

[200-250]km-[500-550]km

Median [100-150]km [200-250]km [150-200]km [300-350]km

Average [100-150]km [200-250]km [150-200]km [350-400]km

(181) Third, as set out in Table 3, the median and average delivery radii of the plants of the Parties' main competitors in terms of road distance indicate that 90% of the sales of the Parties' main competitors fall within a 250km circular catchment area based on geodesic distance.

Table 3 - Range, median and average of the delivery distances of the competitors’ plants

70% of sales – 90% of sales – road distance road distance

Range 125km-260km 175km-375km

Median 200km 300km

Average 200km 297km

(182) Fourth, circular catchment areas representing on average 90% of deliveries around the Parties' plants are more representative in this case than circular catchment areas representing on average 70% of deliveries around the Parties' plants. This is for the following reasons.

(183) In the first place, a circular catchment area representing on average 90% of deliveries around the Parties' plants allows the Commission to identify more readily all the potential competitors that may constrain the Parties and to compute market shares that reflect more adequately the number and strength of those competitors. By contrast, calculating market shares on the basis of a circular catchment area representing on average 70% of deliveries around Cemex's plant in Split leads to the exclusion from the calculation of Titan, even though it is active in the circular catchment area around the Split plant.

(184) In the second place, a circular catchment area representing on average 90% of deliveries around the Parties' plants allows the Commission to include in its assessment a higher share of the sales of Cemex's Split plant which, [strategic information].

6.3.3.3. Territories outside the EEA but within the circular catchment areas around the Parties' plants

(185) For the reasons set out in recitals (186) to (188), the Commission concludes that it can be left open whether parts of the circular catchment areas around the Parties' plants but not falling within the EEA, in particular Bosnia-Herzegovina, should be included in the relevant markets. This is because the Transaction will significantly impede effective competition under both alternative market definitions.

(186) There are arguments both for and against including parts of the circular catchment areas around the Parties' plants but not falling within the EEA within the relevant markets.

(187) On the one hand, there is trade of grey cement across the Croatian-Bosnian border which has increased over the last years. Import volumes from Bosnia-Herzegovina to Croatia (at around 100kT per year) and from Croatia to Bosnia-Herzegovina are quite significant which suggests that the border as such does not inhibit trade of grey cement between those countries.

(188) On the other hand, the lack of sufficient infrastructure seems to be an obstacle to timely deliveries of grey cement from Bosnia-Herzegovina to Croatia if the importer has to travel long distances in Bosnia-Herzegovina before reaching Croatia, as Asamer and Titan have to do, especially considering Bosnia-Herzegovina's mountainous territory. The construction of a highway connecting Bosnia-Herzegovina to Dalmatia is currently under way. However the finalisation of the works will take at least three to five more years. Therefore, transport through Bosnia-Herzegovina is less time and cost effective than transport through Croatia which affects in particular the competitiveness of potential importers located further away from the Croatian border such as Asamer (with respect to southern and western Croatia) and Titan.

6.3.3.4. Modified catchment areas based on road delivery distances and delivery patterns

(189) For the reasons set out in recitals (191) to (206), the Commission concludes that it can be left open whether the 250km circular catchment area around the Parties' plants needs to be subject to further refinements to reflect the specific delivery distances to individual customers and the actual road network conditions in different parts of the circular catchment area. This is because the Transaction will significantly impede effective competition under both alternative market definitions.

(190) There are arguments in favour of further refining the 250km circular catchment area around the Parties' plants to reflect the specific delivery distances to individual customers and the actual road network conditions in different parts of the circular catchment area.

(191) First, the circular 250km catchment areas around the Parties' plants are based on the fact that the conditions for deliveries to the customers in those areas are sufficiently homogeneous and can be distinguished from neighbouring areas. As explained in section 7.4, however, there are variations in competitive conditions within the circular catchment area, mainly due to topographical factors.

(192) Second, the real driving distance to reach customers in Slavonia located within the 250km circular catchment area is longer than for customers on the Adriatic coast or in central Croatia. As transport costs are high compared to the value of grey cement, deliveries are typically done using the shortest possible route. To deliver grey cement from Dalmatia to Slavonia, or vice versa, the shortest route would lead through Bosnia-Herzegovina. However, for such deliveries within Croatia, deliveries do not transit through Bosnia-Herzegovina but rather take a longer route through Croatia (as shown in Figure 4).

ID2111, minutes of a telephone interview with Asamer of 30 November 2016.

ID2156, minutes of a telephone interview with Titan of 29 November 2016. ID2111, minutes of a telephone interview with Asamer of 30 November 2016.

Figure 4: 250km circular catchment area around Split and actual minimum driving distances to areas located on the 250km perimeter

Source: Form CO, Google Maps, the Commission

(193) The Commission has therefore examined a modified 250km circular catchment area for the Split plant based on real delivery distances within the 250km radius around Split.

(194) In a first step, the Commission determined the average road distance to the customers on the outer 250km rim with relatively straight/linear road connections from Split such as, for example, in the Rijeka and Zagreb urban areas.

(195) In a second step, the Commission applied that average distance to the actual road network in Croatia to delineate the area where the customers can be served from Split up to that average distance, assuming no transits through Bosnia-Herzegovina.

(196) Based on this approach, the Notifying Parties calculated that the average road distance to supply a customer around the Zagreb area was […]km. On that basis, a […]km catchment area around the Split plant was drawn up.

ID1397, M.7878_Form CO_Response to Art 11 RFI_20161121.pdf, p. 6.

(197) The modified Split catchment area is corroborated by the Commission’s analysis of actual deliveries from the Cemex plant in Split accounting for 90% of cement sales. Figure 6 indicates [sales data]. That analysis suggests that the 250 km geodesic radius overstates the catchment area of the Cemex Split plant because [sales data] which is due to the topography of Croatia and of the neighbouring countries.

Figure 6: Overlap of […]km catchment area around Split, 250km circular catchment area around Split and customer map for 90% of sales from Split (customers represented as black dots)

ID1397, M.7878_Form CO_Response to Art 11 RFI_20161121.pdf, Figures 10 and 12.

(198) Such modification is not necessary for the circular catchment area around the Kakanj plant. The plant can use relatively straight road connections between Bosnia-Herzegovina and Croatia so that applying the modification approach to the Kakanj plant is not expected to change the results of the Commission's assessment.

(199) HeidelbergCement contested this market definition in its reply to the Statement of Objections and claimed that the Commission’s assumption that cement does not transit through Bosnia-Herzegovina was flawed. According to HeidelbergCement, “cement can and does transit through Bosnia with Croatian start- and endpoints” and Nexe could easily save three hours driving time for deliveries from Slavonia to Dalmatia by transporting cement through Bosnia-Herzegovina. The Commission disagrees with HeidelbergCement's claims for the following reasons.

(200) First, the Parties provide no evidence of such transit, which would be in the particular interest for two suppliers in Croatia: (1) for Cemex Croatia located in Dalmatia to access customers in Slavonia-Continental East, and (2) for Nexe located in Slavonia-Continental East to access customers in Dalmatia. [Strategic information].

(201) In the first place, and as also apparent from Figure 6, [strategic information].

(202) In the second place, Nexe transports (very limited volumes of) cement to Dalmatia using the Croatian road network circumventing Bosnia-Herzegovina. Moreover, it considers that expansion to Dalmatia is only possible through use of the Croatian railway network (if and when appropriate terminal infrastructure in Dalmatia is secured) and not through road deliveries transiting Bosnia-Herzegovina.

(203) Second, Cemex itself confirmed that “[content of internal documents]“.

(204) Third, Cemex confirmed that the fees for transit through Bosnia-Herzegovina could be considerably higher than for regular imports: “With regular customs procedures without any preferential status, forwarding agent fee for export is approximately […] EUR per delivery and additional […] EUR per delivery for import in Bosnia. [Strategic information]. But in that situation the forwarding agent fee would be […] EUR per delivery ([…] EUR export from Croatia, […] EUR for transit in Bosnia and […] EUR to import again in Croatia).” This would amount to an additional cost of around […] EUR/t, or approximately […]% of the final price.

(205) Fourth, information provided by Nexe suggests that while transit through Bosnia-Herzegovina is possible, it has several disadvantages compared to transport through Croatia. On the one hand, for the use of Croatian hauliers Bosnian work permits would be needed for the drivers. On the other hand, the use of Bosnian hauliers is restricted by the Croatian rules on cabotage, which govern the conditions of transport by foreign hauliers of cargo between a start- and endpoint in Croatia and do not allow the use of non-EEA hauliers, including Bosnian hauliers.

(206) Fifth, the Parties have not provided evidence for their assertion that three hours of driving time could be saved by transporting cement through Bosnia-Herzegovina. Publicly available sources suggest that while delivery distances may be higher for transports through Croatia, driving times are in fact comparable (for both routes, around 8 hours from Našice to Split for heavy lorries). In addition, the overall travel time for transits through Bosnia-Herzegovina would include the waiting times at two border crossings, and could thus in fact be significantly longer than for cement transports through Croatia.

(207) Sixth, the example of Titan crossing Bosnia-Herzegovina for its deliveries of bagged cement from Serbia to Dalmatia is irrelevant for the assessment whether cement transits Bosnia-Herzegovina for deliveries between Dalmatia and Slavonia.

7. COMPETITIVE ASSESSMENT

7.1. Principles

(208) Article 2 of the Merger Regulation provides that the Commission has to appraise concentrations within the scope of the Merger Regulation with a view to establishing whether or not they are compatible with the internal market. For that purpose, the Commission must assess, pursuant to Article 2(2) and (3), whether or not a concentration would significantly impede effective competition, in particular as a result of the creation or strengthening of a dominant position, in the internal market or a substantial part of it. Accordingly, the Commission must take into account any significant impediment to effective competition likely to be caused by a concentration, in particular the creation or strengthening of a dominant position.

(209) Generally, a merger giving rise to 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.

(210) 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.

(211) The overall concentration level in a market may also provide useful information about the competitive situation. In order to measure concentration levels, the Commission often applies the Herfindahl-Hirschman Index ('HHI'). The HHI is calculated by summing the squares of the individual market shares of all the firms in the market. While the absolute level of the HHI can give an initial indication of the competitive pressure in a market post-merger, the change in the HHI (known as the 'delta') is a useful proxy for the change in concentration directly brought about by a merger.

(212) 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.

(213) In assessing the competitive effects of a merger, the Commission compares the competitive conditions that would result from the notified merger with the conditions that would have prevailed without the merger.

(214) 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.

(215) 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.

(216) In evaluating the likelihood of non-coordinated effects potentially caused by a merger, it is important to assess to what extent the products of one merging party are close substitutes to the products sold by the other merging party. 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.

(217) 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, that may be the case for customers that have used dual sourcing from the two merging firms as a means of obtaining competitive prices.

(218) 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. 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 those firms have the incentive to react aggressively to a post-merger price increase.

(219) 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.

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

(221) 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.

7.2. Application to this case

(222) The Commission concludes that the Transaction will significantly impede effective competition through non-coordinated effects, which could amount in particular to the creation of a dominant position, in grey cement markets alternatively defined as (i) a 250km circular catchment area around Cemex's Split plant or (ii) a modified 250km catchment area around Cemex's Split plant. Both of those markets include the overlap lens with the 250km catchment area of the Kakanj plant.

(223) This conclusion is based on the following elements:

(a) the combined market shares of the Parties and the market share increments in each alternative catchment area will be high.

(b) the Parties are close competitors because DDC's plant in Kakanj/Bosnia-Herzegovina is geographically the closest plant to Cemex in Split/Croatia and DDC has been aggressively targeting Cemex's customers.

(c) DDC (and to a more limited degree Italcementi) is currently pursuing a [strategic information] strategy for Croatia, and [strategic information] whereas other producers, [strategic information], are reducing output.

(d) domestic suppliers and importers will not sufficiently constrain the Parties due to geographic distance and other factors such as security of supply, a lower market acceptance and reduced ability to engage into barter trading and to assess the creditworthiness of customers.

(e) there are no potential competitors whose market entry would be sufficiently likely, timely and sufficient.

(f) the Transaction is likely to result in quantifiable price increases.

(224) A more detailed analysis of those elements is provided in sections 7.5 to 7.12. Before assessing those elements, the Commission first provides an outline of the main suppliers of grey cement in the relevant catchment areas (section 7.3) and of the geographic differentiation within the catchment areas (section 7.4).

167Horizontal Merger Guidelines, paragraph 36.

168Horizontal Merger Guidelines, paragraph 60.

169Horizontal Merger Guidelines, paragraph 68.

7.3. Main suppliers of grey cement in the relevant catchment areas

(225) Cemex Croatia is the largest cement supplier in Croatia. HeidelbergCement (via DDC’s plants in Kakanj/Bosnia-Herzegovina, and Beremend/Hungary and via Italcementi’s plant in Trieste/Italy) has been by far the largest importer into Croatia, alone accounting for […]kt of imports, or […]% of the overall import volume ([…]kt in 2015).

(226) Apart from the Parties, the main suppliers of grey cement in the relevant catchment areas are the following:

(a) LafargeHolcim, active in Croatia through its subsidiary Holcim (Hrvatska) d.o.o. Koromačno, which operates one integrated grey cement plant in Koromačno (western coastal Croatia) and two grey cement storage terminals, one in Jastrebarsko (Zagreb) and one in Zadar;

(b) Nexe group, a local supplier headquartered in Našice where it operates one grey cement plant;

(c) Asamer, present in Croatia through cement imports from its production plant in Lukavac in Bosnia-Herzegovina;

(d) Titan, active in the import of cement into Croatia from its production plant located in Kosjerić, Serbia;

(e) W&P Zement group ("W&P"), an Austrian cement producer operating, among others, a plant in Anhovo, Slovenia; and

(f) Colacem, an Italian company active in the production of cement and having production plants in Italy and Albania.

170ID444, Form CO, paragraph 167.

(227) The Parties also claim that there are a number of other smaller actual or potential suppliers in the relevant catchment areas:

(a) Cementizillo is an Italian family-owned company operating two cement production plants, one in Monselice (Veneto region) and another one in Fanna (Friuli-Venezia Giulia).

(b) Sacci is an important cement player of the cement sector in Italy, operating five cement production plants in the Central and Northern parts of the country. On 29 July 2016, Cementir Holding acquired Sacci’s cement and ready-mixed concrete business division.

(c) Cementir Holding is an Italian multinational company that produces and distributes grey and white cement, ready-mix concrete, aggregates and concrete products. Among others, Cementir operates a cement production plant in Taranto (Apulia).

(d) Gruppo Grigolin is an Italian business group active in the building construction industry with production units in Treviso (Veneto).

(e) CRH is a global diversified building materials group with operations in 31 countries worldwide. CRH currently has no activities in Croatia but is present in two neighbouring countries Hungary and Serbia through wholly owned subsidiaries with operations in cement, ready-mix concrete and aggregates.

(f) Buzzi is active in Italy in the production and distribution of cement, ready-mix concrete, natural aggregates and related products.

(g) Cimsa is a Turkish cement producer carrying out its operations with its 5 integrated plants in Mersin, Eskişehir, Kayseri, Niğde and Afyonkarahisar, a grinding facility in Ankara, Cement Packing facility in Marmara terminal and Cement Packaging facility in Malatya.

(h) Limak Cement Group possesses the third biggest production capacity of Turkey with 10 production plants.

(228) For the reasons explained in section 7.7, the Commission concludes, however, that none of those suppliers will, post-Transaction, have the ability and incentives to expand to sufficiently constrain the merged entity.

7.4. Geographic differentiation within the relevant catchment areas

(229) The Commission concludes that while the relevant catchment areas are sufficiently homogenous and can be distinguished from neighbouring areas, there are variations in competitive conditions within these areas that need to be taken into account in the competitive assessment. This is for the following reasons.

(230) First, both domestic cement suppliers (Cemex, Nexe, LafargeHolcim) and DDC divide Croatia in regional markets to analyse the competitive situation in Croatia in the ordinary course of business due to geographic divisions, logistic costs, and the ensuing difference in competitive conditions. The Parties' own divisions of Croatia into regional markets are set out in Figure 8 and Figure 9.

174Figure 8: Regional markets as used by DDC

[…]

175Figure 9: Regional markets as used by Cemex

[…]

(231) While the subdivisions differ [strategic information], both DDC and Cemex define [strategic information]. Moreover, Cemex consistently refers to Dalmatia as its [content of internal document].

(232) Second, cement suppliers position themselves differently to compete effectively in a given region. For example, Cemex Split sells comparatively [sales data] volumes in Slavonia ([…]kt out of […]kt of Croatia sales, […]%) and Central Croatia ([…]kt out of […]kT, […]%) and focuses on Dalmatia ([…]kt out of […]kt, […]%) in line with the effective driving distance to reach customers. The ensuing geographic differentiation leads to different degrees of concentration within the market which is reflected in market shares based on actual sales in different regions of Croatia.

173See, for example, replies to question 9 to Q-2 – Questionnaire to competitors (ID675; ID683; ID699).

174ID437-103, M.7878_Form CO_Annex 5.4.-I.-(11)_HC Market Study Bosnia and Western Balkan.PDF.

175ID437-40, M.7878_Annex_RFI 20160711_QP7_Sales Nov 2012 c (CX 157).PDF, p. 4.

(233) Third, cement companies price discriminate across regions in Croatia. Figure 10 indicates that Cemex’s delivered cement prices in [pricing information] have been, at least over the last four years, [pricing information] than in other regions in Croatia.

179Figure 10: Cemex average prices in regions in Croatia

[…]

(234) Fourth, the Parties' criticism of the Commission's finding on geographic variations is unfounded and in part inconsistent with the Parties’ own statements. According to the Parties, the Commission’s competitive assessment resulting in the finding of geographic differentiation is in contradiction with its market definition. The case law on market definition “leaves no room for substantial differences of competitive conditions within one and the same relevant geographic market”. In the Parties’ view, the Commission's approach of taking into account geographic differentiation serves “to covertly and indirectly define “micro markets”.

(235) In the first place, the scope of the geographic markets has been defined based on the assessment of facts collected in the in-depth investigation as presented in section 6.3.3. The Parties themselves consider that the scope of the relevant geographic market is broad, notably the circular catchment area of 250km around the Cemex plant in Split, and do not dispute the Commission’s findings in this respect.

(236) In the second place, the Parties acknowledge that “in markets that heavily rely on transport costs geographic markets may be characterised by gradual shifts of demand substitution” and that “some degree of variation may be taken into account”. Moreover, the Parties do not contest the factual findings pointing to the existence of geographic differentiation which includes evidence that the Parties themselves differentiate their competitive analysis for various parts of the relevant catchment areas in the ordinary course of business.

(237) In the third place, the principles of assessing differentiated markets are set out in the Horizontal Merger Guidelines and are standard Commission practice. The Horizontal Merger Guidelines clarify that the same principles established for product differentiation apply to geographic differentiation, which is reflected in the Commission’s decisional practice concerning cement markets: the definition of 150km and 250km catchment areas "cannot be applied statically across all markets. Defining the relevant geographic markets as circles around a grey cement supplier’s plant may lead to the inclusion in the same geographic market of customers facing differing supply conditions, in particular a differing number of close-by supply alternatives. […] Therefore, while taking the overlaps of catchment areas drawn on maps as the main focus of its assessment, the Commission will also take into account the actual supply patterns and the actual number and strength of suppliers available to cement customers in different locations within the catchment areas.”

178For example, the average difference in delivered prices of Cemex’s CEM II 42.5 between [pricing information] and [pricing information] was […]EUR/t, or […]% in 2015 ([…]EUR/t, […]EUR/t).

179ID1078-272, Annex_Art 11 RFI 20161028_Q55.01.

(238) In the fourth place, the Commission applies those principles by assessing the degree of rivalry between the Parties and takes into account that the Transaction eliminates close competitors in terms of customer coverage and commercial strategy (see section 7.6) and that there is a lower degree of substitutability with products from more remote suppliers (see notably sections 7.7.2.2 - 7.7.3). In addition, the evidence set out in recitals (287) to (292) shows that the Transaction will lead to a stronger increase in concentration in Dalmatia, and indicates that the Parties are closer competitors in that area.

(239) In the fifth place, the competitive assessment is not limited to Dalmatia, where the Parties’ overlap is more pronounced. In the entire relevant markets both actual and potential competition between the Parties would be lost after the Transaction, albeit to differing degrees depending on the availability of alternative suppliers. This is reflected in the high concentration levels for the overall market, and the higher market shares of the Parties in Dalmatia.

7.5. Market shares

(240) The Commission concludes that, post-Transaction, the Parties will have high market shares in the catchment areas consisting of (i) the modified catchment area around Split (and its overlap with the 250km circular catchment area around Kakanj) and (ii) the circular catchment area based on a 250km radius around Split (and its overlap area with the 250km circular catchment area around Kakanj).

7.5.1. Relevance for the competitive assessment of market shares and the number of alternative suppliers of grey cement

(241) The Parties argue that market shares are not a useful indicator of market power as customers of grey cement in the relevant markets solicit bids from one or more suppliers, can easily switch supplier, and will continue to be able to source grey cement from, on average, more than four alternative suppliers after the Transaction, each of which has spare capacity and the incentive to supply customers.

(242) Moreover, and in any event, the Commission's assessment should focus on the number of alternative suppliers of grey cement within the relevant markets as it is a more useful indicator than market shares.

185Case M.7744 – HeidelbergCement/Italcementi, recital 40; see also cases M.7252 – Holcim/Lafarge, recitals 61-63; M.7550 – CRH/Holcim Lafarge divestment business, recital 48, M.7054 – Cemex/Holcim Assets, recitals 57-58; M.7009 – Holcim/Cemex West, recitals 69-70.

186Horizontal Merger Guidelines, paragraph 28.

187While the Commission's conclusion is based on market shares in 2015, it has no reason to believe that market shares would have been materially different in 2013 and 2014.

188This Decision does not assess the market shares in the 250km circular catchment area around Kakanj as that catchment area is already covered in the assessment (for the overlaps with the Split circular catchment area), falls outside the Commission's jurisdiction (for the non-EEA territory and Hungary) or is immaterial to the assessment (additional limited parts of Slavonia close to the Hungarian border) see map in Figure 14.

189ID2415, HeidelbergCement reply to the Statement of Objections, Section B.IV.1, and Annex 1, Section 3.2 (ID2410).

(243) For the reasons set out in sections 7.5.1.1 and 7.5.1.2, the Commission concludes that:

(a) market shares are a useful indicator of the competitive importance of both the Parties and their competitors in the relevant markets; and

(b) the number of alternative suppliers of grey cement is not a useful indicator of the competitive importance of both the Parties and their competitors in the relevant markets.

7.5.1.1. Market shares are a useful indicator of the competitive importance of both the Parties and their competitors

(244) There are a number of reasons why market shares provide a useful first indication of the competitive importance of both the Parties and their competitors in the relevant markets.

(245) First, the Parties themselves use market shares [strategic information], as exemplified in Figure 11 for Cemex and in Figure 12 for DDC.

190Figure 11: Cemex’ market share analysis for Croatia

[…]

191Figure 12: DDC's market share analysis for Croatia ([…] the document contains further market share estimates […])

[…]

(246) Second, market shares and HHI levels based on market shares have been used in all past cases concerning the cement industry.

(247) Third, this conclusion follows from a number of characteristics of the relevant markets.

(248) In the first place, geographic differentiation affects the competitive positions of grey cement suppliers in or close to the relevant markets, with suppliers located further away from a customer being at a competitive disadvantage of supplying that customer due to higher transport costs and security of supply concerns. Such differences in competitiveness (which can be reinforced by other differentiating factors such as reputation and quality) will affect the degree of success of a grey cement supplier in acquiring customers in a specific area, and hence will be reflected in sales market shares. The Parties' statement that "[i]f one plant is located much closer to a particular group of customers (as for Dalmatia it is Cemex with its plant in Split), it is expected that in the end this plant (Cemex) will gain a large number of orders in the relevant area (Dalmatia)" is in line with this view.

190ID437-39, Annex_RFI 20160711 QP7_Sales Mar 2013 (CX 164), LOF.

191ID437-103, M.7878_Form CO_Annex 5.4.-I.-(11)_HC Market Study Bosnia and Western Balkan.PDF.

(249) In the second place, sales market shares in grey cement are the outcome of past negotiations between suppliers and a large number of customers over volumes of grey cement which, taken individually, do not amount to a large percentage of total sales in the relevant markets. On the one hand, a high market share of a supplier thus indicates that this supplier has been consistently more successful in attracting business, which in turn indicates that the supplier has a competitive advantage and a degree of market power. On the other hand, a low market share of a supplier indicates that despite a large number of competitive opportunities, it is not a credible alternative for customers and does not exert the same competitive constraint as firms with higher market shares. The Parties' argument that "[…] firms with low (or even no) market share are decisive to limit prices to the level at which they are" is hence incorrect.

194ID2415, HeidelbergCement reply to the Statement of Objections, Section B.IV.1, and Annex 1, para 11 (ID2410).

(250) In the third place, the reasoning in recitals (248) and (249) is not affected by the fact that customers negotiate individual prices with one or more suppliers and can switch accordingly. The Parties' argument that customer specific price negotiations coupled with ease of customer switching would give rise to a "bidding market logic" that would imply that market shares are uninformative is hence incorrect.

(251) Fourth, contrary to the Parties' claims, the conclusion that market shares are a useful first indication of the competitive importance of both the Parties and their competitors in the relevant markets is not affected by the variations in competitive conditions within the relevant markets.

(252) In the first place, the overlap lens of the Split and Kakanj plants as depicted in Figure 13 and Figure 14 accounts for a large majority of sales of grey cement in the modified catchment area (EEA: […]kt out of […]kt, or […]% of cement volumes; all territory: […]kt out of […]kt, or […]% of cement volumes) and in the 250km circular catchment area around Split (EEA: […]kt out of […]kt, or […]% of cement volumes; all territory: […]kt out of […]kt, or […]% of cement volumes).

193ID2415, HeidelbergCement reply to the Statement of Objections, Section B.IV.1, and Annex 1, para 11 (ID2410).

(253) In the second place, the Parties have significant competitive interactions outside the overlap lens, for example in major consumption centres in Zagreb (due to the sales from DDC’s Beremend plant) and Rijeka (due to the sales from Italcementi’s plant in Trieste) at the margins of the relevant catchment areas.

(254) In the third place, there is no clear dividing line within the relevant markets along which the conditions of competition suddenly change. Rather conditions of competition change gradually in the relevant markets, depending on the respective distance of customers from the plants of the Parties and of their competitors.

194ID1142 Response to Commission RFI of 8 November 2016; ID2242 Reply: Clarification about the request for information - Art 11 Request for information to Asamer_final_sent version; ID1185, EC 17112016.docx and ID1842 UPITNIK EK rok-16.11.2016.pdf. Moreover, according to the Parties' analysis, their [sales data] largest customers of grey cement in Dalmatia account for less than […]% of their combined grey cement sales in Dalmatia. The remaining […]% of sales in Dalmatia are made up of a [sales data] customer base (ID2410 HeidelbergCement reply to the Statement of Objections, Section B.IV.1, and Annex 1, Figure 3). Furthermore, [sales data] further indicates that market shares are not affected by the sourcing behaviour of a small number of large customers (ID437-103, M.7878_Form CO_Annex 5.4.-I.-(11)_HC Market Study Bosnia and Western Balkan.PDF).

(255) There are a number of reasons why the number of alternative suppliers of grey cement that customers in the relevant markets can turn to is not a useful indicator of the competitive importance of both the Parties and their competitors.

(256) First, unlike market shares, the number of alternative grey cement suppliers that a customer can allegedly turn to does not reflect differences in competitive positions arising from geographic distance to the customer or other differentiating factors such as reputation and quality.

(257) Second, the maximum effects analysis proposed by the Parties – which uses a simple framework to analyse bidding for individual customers in geographically differentiated relevant markets – indicates that it is not the number of alternative suppliers that will constrain the merged entity but rather their geographic location.

(258) Third, the fact that customers tend to request offers from only a relatively small number of established grey cement suppliers in the relevant markets indicates that suppliers with low or zero market shares that are included in the alternative competitor counts do not exert an effective competitive constraint. If, as argued by the Parties, such suppliers were readily available competitive alternatives that could serve customers even at a discount relative to current prices, then customers should see benefit in contacting additional suppliers in order to achieve lower prices even pre-Transaction.

(259) Fourth, the Parties' competitor counts are, in any event, based on a flawed methodology which likely overstates the number of competitors that have a specific customer within their catchment areas. This is because the Parties' analysis likely over-states actual catchment areas of some plants and terminals as it applies the same geodesic distance of 250km to identify catchment areas, disregarding that catchment areas vary significantly depending on the identity of the supplier and the type of asset (plants, grinding stations, terminal).

7.5.2. Market share methodology

(260) The Commission has analysed the Parties' market share estimates which have been computed based on two main methodologies: (i) sales market shares based on actual sales (and relative capacities for competitors); and (ii) capacity market shares based on production capacities. Both calculations are equally important for the competitive assessment: while capacity shares indicate the extent to which the suppliers could compete in the relevant markets, the sales market shares indicate the extent to which suppliers actually compete in the relevant markets. Market shares have been calculated on a volume basis rather than on a revenue basis due to the better availability of volume data.

197In the Parties' maximum effects analysis, the competitive constraint on the merged entity depends, for each customer, only on the location of the next closest alternative supplier and not on the number of supplier. (ID437-255 Annex 6.III.1.-Grey Cement_HR_[advisor's identity] economic analysis_revised, p. 30-33) The Parties' analysis is useful to illustrate this general points but also has certain limitations as discussed in detail in section 7.12.2.

198See responses to Questionnaire to Customers (Q1) dated 6, 7 and 15 September 2016, question 9.

199ID2410, HeidelbergCement reply to the Statement of Objections, Annex 1, para 49 and section 3.3.3.

See section 6.3.3.2.

(261) The Parties have submitted market shares for the 250km circular catchment areas drawn around their respective plants. At the Commission's request, the Parties have also provided market shares for the overlap lens between the Parties' plants in Split and Kakanj, for the modified Split catchment area based on road transport distances and patterns, as well as for the overlap lens for the modified area and the 250km circular catchment area around Kakanj. For each of those areas, the Parties have provided shares differentiating also by territory (EEA only vs all territory) and type of supplies (bulk, bagged).

7.5.2.1. Methodology of sales market shares submitted by the Parties

(262) For the calculation of sales market shares, the Parties used (i) the estimated total size of the catchment areas; and (ii) sales by the Parties and their competitors in those areas.

(263) The size of the catchment areas was computed by the Parties as the product of (i) cement consumption per capita in the NUTS-3 administrative regions (in Croatia) or in the respective Member State (outside Croatia) and (ii) the population of the catchment area. As regards the population estimates for the catchment areas, the Parties have used population data based on the NUTS-3 administrative regions. In case of only partial overlaps with the modified and circular 250 km catchment areas, the Parties used rounded thresholds (0%, 25%, 50%, 75%, and 100%) to match the catchment area with the NUTS-3 administrative regions more accurately.

(264) While the Parties used their actual sales data to calculate their own market shares, they estimated the sales of competitors based on public information on the capacity of their plants. The Parties also included in their estimates both internal sales and external sales to third parties as no data on internal sales were available for competitors or for the total market. The sales market shares of competitors were estimated by allocating the catchment area volumes minus the Parties’ volumes to each competitor in proportion to its production capacity shares in those areas.

7.5.2.2. Methodology of capacity market shares submitted by the Parties

(265) For the calculation of capacity market shares, the Parties used (i) the estimated total size of the catchment areas; and (ii) the capacities of the Parties and their competitors in those areas.

(266) In relation to both catchment areas, the capacity market shares were calculated by taking into account both plants located within the modified and the 250 km circular catchment areas and plants outside the relevant areas whose catchment areas overlap with the relevant areas. For each competitor, the capacities of all plants that have geodesic overlaps with the relevant catchment areas are summed up. However, the

(267) Market shares were calculated by dividing the sum of the weighted capacities of all sites belonging to the same competitor by the market total, which is the sum of the weighted capacities of all the plants whose catchment areas intersect with the relevant modified or circular catchment areas.

7.5.2.3. The Commission's assessment of Parties' market share methodology

(268) The Commission has used and analysed the Parties' market share estimates. Due to the shortcomings of the Parties' market share estimates as summarized in recital (269), the Commission has complemented the Parties' calculations by reconstructing the actual sales shares for four regions in Croatia and by calculating customer-centred capacity shares based on the average and individual sales radii of the competing plants and terminals. These calculations have been further complemented by the Commission's assessment of other available evidence, such as the qualitative results of the Commission's investigation.

(269) The Commission notes that calculating market shares in the relevant catchment areas has several challenges due to the following considerations which may result in inaccuracies of the calculations provided by the Parties and in underestimating the Parties' combined shares:

(a) as regards sales shares, the Parties have had to rely on assumptions as regards the number of inhabitants and the uniform per capita consumption of cement in the relevant catchment areas. Actual demand patterns may therefore differ from the assumptions made by the Parties.

(b) as in previous cases, the estimates of sales shares are partially based on sales data (for the Parties) and partially on capacity data (for the competitors). The sales share estimates for competitors therefore do not reflect actual sales and may distort the competitive strength of competitors in terms of sales.

(c) the Parties have adjusted downward their own nominal capacity and used effective production capacity instead. While Kakanj capacity is publicly stated as 770kT, the Parties have adjusted that to […]kT. On the other hand, they have not used the actual effective capacity for the competitors, but relied on public figures for the nominal capacity of plants, which were in part adjusted based on the Parties' market knowledge. However, the Commission’s reconstruction of market shares based on effective capacity (which is better suited to determine the actual capacity utilisation rates than the nameplate capacity), indicates that, considering the market as a whole, the Parties’ estimates do not deviate materially from the data of competitors. Therefore, while the Parties' combined capacity appear to have been calculated rather accurately, the individual capacity shares of competitors may be distorted.

(d) the Parties have overstated competitors' capacity by applying the same standard capacity of 250kT for their competitors' cement terminals while using actual capacity for Cemex' own terminals which is [capacity] (Cemex Bakar: […]kT; Cemex Metković: […]kT). Similarly, LafargeHolcim submits that the actual capacity of its terminal in Zadar on the Croatian coast is [65-75]kt.

(e) the Parties have calculated sales market shares for Italcementi based on actual sales in Croatia to a single customer location. However, as that customer, A.R.M.A.C.O.M., is a trader selling Italcementi cement to a multitude of customers in various parts of Croatia, and the Parties do not have visibility on the final destination and volumes of the sales, the Commission considers that the same methodology as for competitors should also be applied to Italcementi. (For a more general discussion of why A.R.M.A.C.O.M. sales should be attributed to Italcementi, see recital (274)).

(f) the Parties used the same 250km geodesic radii for the plants and for the terminals, thus overstating the geographic reach of some terminals. It appears, however, that the terminals should have a shorter radius than the plants because the cost incurred to supply the terminals will necessarily reduce the radius at which the terminal can sell to end customers. If a different radius is assumed for terminals relative to plants, the Parties' combined capacity shares and the calculated HHI levels would increase.

(270) The Commission considers that market shares based on the modified Split catchment area are best suited for the assessment of the Transaction for the following reasons:

(a) they are based on real road distances;

(b) they take into account actual topography, the shape of the Croatian territory and the layout of its road network, as well as real-life patterns of transportation routes. Concerning the latter factor, the approach takes into account the fact that cement suppliers in Croatia do not transit through Bosnia-Herzegovina to serve customers in Croatia even if that may be the shortest route (notably on the north-south axis Slavonia-Dalmatia); and

(c) they reflect better the lower competitive pressure exerted by companies which in fact have to cover significantly longer road distances to supply customers than what the 250km circular catchment area would suggest.

7.5.3. Sales market shares

(271) The Commission has analysed the Parties' estimated market shares for: (i) the relevant market comprising the modified catchment area around the Split plant (and its overlap with the Kakanj 250km circular catchment area); and (ii) the relevant market comprising the circular 250km catchment area around the Split plant (and its overlap with the Kakanj 250km circular catchment area).

(272) The Commission has added together the sales shares of Cemex, DDC and Italcementi to arrive at the Parties' joint market share.

(273) The Commission has attributed to Italcementi the sales of grey cement by […], a third party distributor, for several reasons.

(274) First, this reflects the way that DDC [sales data and strategic information]. Second, Italcementi can control the sales of A.R.M.A.C.O.M. in Croatia by reducing or discontinuing its supplies to that distributor. Third, Italcementi has a close business relationship with A.R.M.A.C.O.M. and […] in overcoming Cemex Croatia’s attempt to block imports of Italcementi’s cement into Croatia, including by initiating legal action against Cemex.

7.5.3.1. Sales market shares based on the modified Split catchment area

(275) The combined sales market shares of the Parties, which exceed [50-60]%, indicate, in combination with other elements assessed in section 7 of this Decision, that the merged entity will enjoy a strong, if not a dominant, position, in the modified Split catchment area.

(276) Table 4 shows the Parties' sales market shares for the modified Split catchment area and for its overlap lens with the 250km circular catchment area in Kakanj in 2015 (the overlap lenses are set out in Figure 13 [EEA territory only] and Figure 14 [all territory]).

Figure 13: overlap between the modified Split catchment area and the 250km catchment area around Kakanj (EEA only)

[…]

Figure 14: overlap between the modified Split catchment area and the 250km circular catchment area around Kakanj (all territory)

[…]

Table 4 - Sales market shares in 2015 for the modified Split catchment area and for the overlap lens between the modified Split catchment area and the Kakanj 250km circular catchment area

Overlap lens Modified Overlap Split lens catchment area

Company Modified Split catchment area

All cement All cement Bag Bulk All cement All cement

EEA EEA EEA EAA all territory all territory

HeidelbergCement [10-20]% [10-20]% [5-10]% [10-20]% [20-30]% [20-30%] / SchwenkDDC

HeidelbergCement [0-5]% [0-5]% [0-5]% [0-5]% [0-5]% [0-5]% (Italcementi)

Cemex [40-50]% [40-50]% [40-50]% [40-50]% [20-30]% [30-40]%

Combined [50-60]% [60-70]% [50-60]% [60-70]% [50-60]% [50-60]%

LafargeHolcim [10-20]% [10-20]% [10-20]% [10-20]% [10-20]% [10-20]%

Asamer [5-10]% [5-10]% [5-10]% [5-10]% [5-10]% [5-10]%

Nexe [5-10]% [5-10]% [5-10]% [5-10]% [5-10]% [5-10]%

W&P [10-20]% [5-10]% [5-10]% [5-10]% [0-5]% [0-5]%

Colacem -- -- -- -- [0-5]% [0-5]%

Titan [0-5]% [0-5]% [0-5]% [0-5]% [5-10]% [5-10]%

Others [5-10]% [0-5]% [0-5]% [0-5]% [5-10]% [5-10]%

Total 100% 100% 100% 100% 100% 100%

Market size (in kT) […] […] […] […] […] […]

(277) In the modified Split 250km catchment area, the Parties accounted for [50-60]% of sales in 2015 in the area excluding non-EEA territory (Cemex: [40-50]%, DDC: [10-20]%, ITC: [0-5]%) and [50-60]% of sales in the area including non-EEA territory (Cemex: [20-30]%, DDC: [20-30]%, ITC: [0-5%]). Market shares were even higher in the bulk cement segment both in the area excluding non-EEA territory (Cemex: [40-50]%; DDC; [10-20]%; ITC: [0-5]%) and including non-EEA territory

(Cemex: [20-30]%; DDC: [20-30]%; ITC: [0-5]%) which represented around two thirds of consumption in Croatia.

(278) In the overlap lens between the modified Split catchment area and the 250km circular catchment area around Kakanj, the Parties accounted for [60-70]% in the area excluding non-EEA territory (Cemex: [40-50]%, DDC: [10-20]%, ITC: [0-5]%) and [50-60]% in the area including non-EEA territory (Cemex: [30-40]%, DDC: [20-30]%, ITC: [0-5]%). The largest remaining competitor, LafargeHolcim, accounted for only [10-20]% while Nexe, the remaining domestic producer, accounted for [5-10]%. In the overlap area (EEA only), the Parties were relatively stronger in bulk cement ([60-70]%) compared to bagged cement ([50-60%]) although market shares for both those potential sub-segments were high.

7.5.3.2. Sales market shares based on 250km radius around Split

(279) The combined sales market shares of the Parties, which exceed [40-50]%, indicate, in combination with other elements assessed in section 7 of this Decision, that the merged entity will enjoy a strong, if not a dominant, position, in the 250km Split circular catchment area.

(280) Table 5 and Table 6 show the Parties' sales market shares in 2015 for the catchment areas around the Split plant and the overlap lens between the Kakanj and the Split plants. Table 5 shows market shares in which Bosnia-Herzegovina and other non-EEA territory has been excluded whereas Table 6 shows figures including all territory, whether EEA or not.

Table 5 - Sales market shares in 2015 for the 250km Split circular catchment area and for the overlap lens between Kakanj and Split (EEA only)

Company Catchment 250 km catchment Overlap area around Split lens

All cement Bag Bulk All cement

HeidelbergCement/SchwenkDDC [10-20]% [5-10]% [10-20]% [10-20]%

HeidelbergCement (Italcementi) [0-5]% [0-5]% [0-5]% [0-5]%

Cemex [30-40]% [30-40]% [30-40]% [30-40]%

Combined [40-50]% [40-50]% [40-50]% [50-60]%

LafargeHolcim [20-30]% [20-30]% [20-30]% [20-30]%

Asamer [5-10]% [5-10]% [5-10]% [5-10]%

Nexe [5-10]% [5-10]% [5-10]% [5-10]%

W&P [10-20]% [10-20]% [10-20]% [5-10]%

Colacem -- -- -- [0-5]%

Titan [0-5]% [0-5]% [0-5]% [0-5]%

Others [5-10]% [5-10]% [5-10]% [0-5]%

Total 100% 100% 100 100%

Market size (in kT) […] […] […] […]

Table 6 - Sales market shares in 2015 for the 250km Split area and for the overlap lens between Kakanj and Split (all territory)

Company 250km catchment area 250 km catchment area around Split lens

All cement Bag Bulk All cement

HeidelbergCement/SchwenkDDC [20-30]% [10-20]% [20-30]% [20-30]%

HeidelbergCement (Italcementi) [0-5]% [0-5]% [0-5]% [0-5]%

Cemex [20-30]% [20-30]% [20-30]% [20-30]%

Combined [40-50]% [40-50]% [50-60]% [40-50]%

LafargeHolcim [10-20]% [10-20]% [10-20]% [10-20]%

Asamer [5-10]% [5-10]% [5-10]% [5-10]%

Nexe [5-10]% [5-10]% [5-10]% [5-10]%

W&P [5-10]% [5-10]% [5-10]% [0-5]%

Colacem [0-5]% [0-5]% [0-5]% [0-5]%

Titan [5-10]% [5-10]% [5-10]% [5-10]%

Others [5-10]% [10-20]% [5-10]% [5-10]%

Total 100% 100% 100% 100%

(281) In the Split 250km circular catchment area, the Parties accounted for [40-50]% to [40-50]% of sales in 2015 (all territory: Cemex: [20-30]%, HeidelbergCement/DDC ("HC/DDC"): [20-30]%, ITC: [0-5]%; EEA only: [40-50]%, Cemex: [30-40]% HC/DDC: [10-20]%, ITC: [0-5]%) whereas in the Split-Kakanj overlap area, they accounted for [40-50]% to [50-60]% of sales in 2015 (all territory: [40-50]%, Cemex: [20-30]%, HC/DDC: [20-30]%, ITC: [0-5]%; EEA only: [50-60]%, Cemex: [30-40]%, HC/DDC: [10-20]%, ITC: [0-5]%). The largest remaining competitor, LafargeHolcim, accounted for [10-20]% to [20-30]% (Split: [10-20]%, overlap lens: [10-20]%, EEA only: Split and the overlap lens: [20-30]%), whereas Nexe, the remaining domestic producer, accounted for only [5-10]%. The Parties were relatively stronger in bulk cement ([40-50]-[50-60]%) compared to bagged cement ([40-50]%) in the Split 250km circular catchment area although market shares for both those potential sub-segments were high.

7.5.4. Capacity market shares

(282) The Commission has also analysed the Parties' estimated capacity market shares for: (i) the relevant market comprising the modified catchment area around Split (and

221 The Commission notes that its theory of harm does not rely on competitors' having insufficient capacity to replace the competitive constraint lost by the Parties, but rather that competitors are at a competitive

its overlap area with the 250km circular catchment area round Kakanj); and (ii) the relevant market comprising the circular 250km catchment area around the Split plant (and its overlap area with the 250km circular catchment area round Kakanj).

7.5.4.1. Capacity market shares based on the modified Split catchment area

(283) Table 7 shows capacity market shares in 2015 for the modified catchment area around Split, and its overlap lens with the 250km Kakanj circular catchment area (EEA only and all territory).

Table 7 - Capacity market shares in 2015 for the modified Split catchment area and its overlap lens with the 250km Kakanj circular catchment area (EEA and all territory)

Company Modified catchment area around Split

Overlap lens modified Split/ Kakanj 250km

222 224 All AllEEA EEA 223 225 territory territory

HeidelbergCement/SchwenkDDC [5-10]% [10-20]% [10-20]% [10-20]%

HeidelbergCement (Italcementi) [0-5]% [0-5]% [0-5]% [0-5]%

Cemex [30-40]% [30-40]% [30-40]% [30-40]%

Combined [40-50]% [40-50]% [40-50]% [40-50]%

LafargeHolcim [20-30]% [10-20]% [10-20]% [10-20]%

Asamer [5-10]% [5-10]% [5-10]% [5-10]%

Nexe [5-10]% [5-10]% [5-10]% [5-10]%

W&P [10-20]% [5-10]% [0-5]% [0-5]%

Colacem -- [0-5]% [0-5]% [0-5]%

Titan [0-5]% [5-10]% [5-10]% [5-10]%

Others [5-10]% [5-10]% [5-10]% [10-20]%

Total 100% 100% 100% 100%

Market size (in kT) […] […] […] […]

(284) The Parties' combined capacity market share in 2015 was [40-50]% to [40-50]% in the modified catchment area as well as the overlap lens irrespective of whether or not the area is limited to the EEA (Cemex: [30-40]%, DDC: [5-10]-[10-20]%, ITC: [0-5]%). The largest remaining competitor, LafargeHolcim, accounted for [10-20]-[20-30]%, whereas Nexe, the remaining domestic producer, accounted for [5-10]%.

disadvantage due to, in particular, their geographic location, see section 0. Capacity shares should therefore be considered mainly as complementing the more relevant analysis based on actual sales shares.

221 ID2073; M.7878_Annex_RFI 20161201.02_01_359km catchment area_capacity shares_excl. non-Croatian territory.

223 ID2075; M.7878_Annex_RFI 20161201.02_03_359km catchment area_capacity shares_incl. non-Croatian territory.

224 ID2074; M.7878_Annex_RFI 20161201.02_02_359km overlap lens_capacity shares_excl. non-Croatian territory.

225 ID2076; M.7878_Annex_RFI 20161201.02_04_359km overlap lens_capacity shares_incl. non-Croatian territory.

7.5.4.2. Capacity market shares based on 250km radius around Split

(285) Table 8 shows capacity market shares in 2015 for the 250km circular catchment area around the Split plant and its overlap lens with the 250km Kakanj circular catchment area (EEA only and all territory).

Table 8 - Capacity market shares in 2015 for the 250km circular Split catchment area and its overlap lens with the 250km Kakanj circular catchment area (EEA and all territory)

Company 250km catchment area around Split Overlap lens Split/ Kakanj 250km

226 228 All AllEEA EEA 227 229 territory territory

HeidelbergCement/SchwenkDDC [5-10]% [10-20]% [10-20]% [10-20]%

HeidelbergCement (Italcementi) [0-5]% [0-5]% [0-5]% [0-5]%

Cemex [30-40]% [30-40]% [30-40]% [30-40]%

Combined [40-50]% [40-50]% [50-60]% [40-50]%

LafargeHolcim [20-30]% [10-20]% [20-30]% [10-20]%

Asamer [5-10]% [5-10]% [5-10]% [5-10]%

Nexe [5-10]% [5-10]% [5-10]% [5-10]%

W&P [10-20]% [5-10]% [5-10]% [0-5]%

Colacem -- [0-5]% -- [0-5]%

Titan [0-5]% [5-10]% [0-5]% [5-10]%

Others [5-10]% [5-10]% [0-5]% [5-10]%

Total 100% 100% 100% 100%

Market size (in kT) […] […] […] […]

(286) The Parties' combined capacity market share in 2015 was [40-50]% to [50-60]% in the 250km circular catchment area and the overlap lens. The largest remaining competitor, LafargeHolcim, accounted for [10-20]-[20-30]%, whereas Nexe, the remaining domestic producer, accounted for [5-10]%.

7.5.5. Market shares and geographic differentiation

(287) Because of geographic differentiation within the relevant catchment areas, the market share of the merged entity in the southern region of Croatia, Dalmatia, is higher than overall in the relevant catchment areas.

(288) Table 9 presents the sales market shares by region computed using the Parties' estimates. In the region of Dalmatia, the merged entity accounted for a sales market share in 2015 of [70-80]% whereas LafargeHolcim accounted for [10-20]%, and Titan for [5-10]%.

226 ID913, M.7878_Annex RFI 20161005_Q03.17_Grey Cement_capacity shares_plants_EEA only_no public ports_no over.PDF.

227 ID911, M.7878_Annex RFI 20161005_Q03.15_Grey Cement_capacity shares_plants_no public ports_no oversea suppl.PDF.

228 ID906, M.7878_Annex RFI 20161005_Q03.10_Grey Cement_capacity shares_250 km overlap lens_EEA only_no public .PDF.

229 ID905, M.7878_Annex RFI 20161005_Q03.09_Grey Cement_capacity shares_250 km overlap lens_no public ports_no.pdf.

lens in Croatia, measured in pre-Transaction Herfindahl-Hirschman-Index ("HHI") levels, post-Transaction HHI levels and differences in HHI levels brought about by the Transaction ("HHI deltas").

Figure 15: Pre-Transaction HHI in the 250km overlap lens, EEA territory only

232 ID2175, Memo in response to RFI of 2 December 2016.PDF.

Figure 16: Post-Transaction HHI in the 250km overlap lens, EEA territory only

Figure 17: Delta in HHI for the 250km overlap lens, EEA territory only

(291) The pre-Transaction HHI levels range between below 2000 in the North of the overlap lens (Central Croatia) and 6501 – 7000 in the centre and South-West of the overlap lens. The post-Transaction HHI levels reach between 2 001 – 2 500 and 8501 – 9 000. The HHI delta ranges between below 250 in the very North-West of the overlap lens to up to 1 751 – 2 000 in the centre and South of the overlap lens.

(292) Similarly, the merged entity’s capacity shares computed based on customer locations show that the combined capacity market shares are high and that the Parties' combined market shares are particularly high in Dalmatia. Based on the data collected from the Parties and their competitors, the Commission calculated capacity shares for each of the Parties’ customer locations by including the capacities of all plants and terminals for which that customer is within the 90% delivery distance. The details of the analysis are presented in the Annex.Figure 18 shows the merged entity's capacity shares using the plant-specific 90% sales radii for each plant/terminal.

Figure 18: Merged entity's capacity market for the overlap customers using plant-specific 90% sales radii

[…]

7.6. Competitive pressure exerted by DDC and Italcementi on Cemex before the Transaction

(293) For the reasons set out in sections 7.6.1 and 7.6.2, the Commission concludes that DDC is an important competitive force in Croatia, in particular in Dalmatia, and a close competitor of Cemex. The Commission further concludes that HeidelbergCement, through Italcementi, is also an important competitive force in western Croatia, albeit to a lesser extent (section 7.6.3).

7.6.1. DDC is an important competitive force in Croatia

(294) DDC is an important competitive force in Croatia because [strategic information].

(295) Expansion by DDC on the Croatian market has been considered amongst the "[content of internal document]". In the period 2013-2015, which coincides with DDC’s [strategic information], DDC’s overall imports to Croatia grew by […]% (from […]kt in 2013 to […]kt in 2015). That growth trend can also be observed for DDC imports to Dalmatia (up […]%: […]kt in 2013, […]kt in 2015). The [strategic information] growth of DDC imports was also noted and analysed in an internal document of Cemex, recording by far the highest growth amongst all importers both in relative and absolute terms. That document also suggests that DDC imports were the driver behind the 6% growth of imports into Croatia in 2014-2015.

(296) TCK's annual report indicates that the source of that expansion was the intensification of competition with Cemex: "Responding to an aggressive presence of competitors from Croatia, Tvornica cementa Kakanj increased its sales in Croatia market. In 2015 export to Croatia was around 70 thousand tons, i.e. Tvornica cementa Kakanj sold 16% of its total volumes in Croatia." While the document

The capacity shares presented in recital (292) are a proxy of the relative competitive strength of different suppliers measured by their relative capacities. See also footnote 221.

Part of the central region of Croatia, the eastern part of the Zagreb area, also displays sizeable market shares for the merged entity. In the Commission's view, however, this is likely due to the fact that certain plants in countries north to Croatia (e.g. Hungary) were not taken into account in the analysis.

The Parties were able to review the Commission's analysis during the Quantitative Data Room procedure described in recital (8).

The Commission also applied a sensitivity scenario using the average 90% radius to, respectively, all plants and all terminals. The results of that sensitivity test confirm the significant capacity shares for the merged entity, in particular in the region of Dalmatia. The details of that sensitivity analysis can be found in the Annex.

ID1088-550, GL_C001_00008657.ppt, p. 66.

ID437-33, M.7878_Annex_RFI 20160711_QP7_Commercial update Sep 2015 (CX 174).PDF, p. 8.

ID1088-1855, GL_C001_00041939.pdf.

59

does not make an explicit reference to Cemex, the overwhelming majority of TCK’s export volumes were sold in Dalmatia, the home sales area of Cemex, and Cemex has in turn also been an important exporter to Bosnia-Herzegovina (see Figure 20).

(297) Moreover, Cemex's internal analysis of market share developments finds that [content of internal document]. To off-set a decline in demand, Heidelberg/DDC was increasing its market share at their expense.

Figure 19: Cemex document Regional Supply_20130930, CX163, ID161-153, p. 6

[…]

(298) The analysis shows that, while on the one hand, certain large operators were reducing their sales volumes in the broader region (Cemex: […]kt, Holcim: […]kt, Titan: […]kt), DDC on the other hand managed to increase its sales volumes by […]kt. Even if the overview concerns the broader region, it remains informative of DDC’s competitive strategy in Croatia, as the data only relates to [strategic information] DDC plants in the region, [strategic information] out of which ([strategic information]), are the most significant source of imports to Croatia. The market share growth trend is also consistent with the [strategic information] of DDC’s import volumes to Croatia (see recital (295)).

(299) By contrast, the Transaction would see DDC transform from an expanding importer in Croatia into the parent company of Cemex Croatia, the largest Croatian incumbent, and customers could no longer benefit from the competitive pressure from those imports. That is confirmed by HeidelbergCement / DDC’s integration plans as presented in Figure 20. Those plans envisage sales reallocation between the plants in Split and Kakanj (referred to as logistics optimisation), and project cement prices to increase, which demonstrates in unambiguous terms that the imports to Croatia would not be sustained after the Transaction.

Figure 20: HeidelbergCement’s integration plans for the Transaction

[…]

7.6.2. DDC is a close competitor of Cemex Croatia

(300) Contrary to the Parties' claim, DDC is a close competitor of Cemex Croatia. This is confirmed by a number of factors.

(301) First, DDC's plant in Kakanj/Bosnia-Herzegovina is geographically the closest cement plant to Cemex's plant in Split. A HeidelbergCement document dedicated to the preparation of the Transaction describes Cemex’s plants in Split as [content of

internal document]. The Parties’ activities overlap throughout the entire Dalmatia and beyond (notably, the Parties appear to be the closest competitors in the south of Bosnia-Herzegovina).

(302) Second, the Parties are each other's closest competitors south-east of Split, where LafargeHolcim is active to a limited extent with only a few customers: “From the Zadar terminal it is not possible to supply the region from Split to Dubrovnik due to high transportation costs. Our only option for supplying the region from Split to Dubrovnik is to use terminals, located either in that region or in Montenegro. Today, only Cemex has terminals in these locations”. That is corroborated by Cemex' internal documents, which suggest that [content of internal document].

(303) Third, DDC is a close competitor of Cemex in Dalmatia, which is Cemex's core market in Croatia, where around […]% of Cemex’s sales in Croatia are realised. This is confirmed by the following.

(304) In the first place, apart from Cemex, in Dalmatia only LafargeHolcim and DDC are selling significant volumes of cement, and in particular in bulk.

(305) In the second place, LafargeHolcim indicated that while imports have become more significant in Croatia in recent years, the Dalmatia region is least influenced by that trend where Cemex's high market shares are only complemented by DDC's imports from Bosnia-Herzegovina and some minor volumes from Titan and from Italcementi (now part of HeidelbergCement).

(306) In the third place, a slide entitled “[content of internal document]” in a Cemex internal document observed that HeidelbergCement’s imports from Bosnia-Herzegovina to Croatia were at stable volumes, however [content of internal document].

(307) In the fourth place, another Cemex document identifies [content of internal documents] as Cemex’ main competitors in Croatia. The same document qualifies Heidelberg as a [content of internal documents] due to its [internal document], notably in Dalmatia: [content of internal documents]. According to Cemex, DDC’s exports into Dalmatia served to [content of internal documents].

Figure 21: excerpt from Cemex document Balkans Strategy 2014

[…]

(308) In the fifth place, internal DDC documents indicate that [content of internal documents and strategic information], as exemplified by the following citations:

(a) “[Content of internal documents].”

(b) “[Content of internal documents].”

(c) “[Content of internal documents].”

(d) “[Content of internal documents].”

(e) “[Content of internal documents].”

(f) "[Content of internal documents]."

(g) "[Content of internal documents]."

(h) "[Content of internal documents]" , […].

(309) That is reflected, for instance, in the statements of customer Strabag: "Heidelbergcement (Duna Drava Cement) contacted us and offered us in year 2013 for the region of Dalmatia competitive conditions for buying and selling slightly better than the market prices of grey cement" and "Yes, in region of Dalmatia Heidelbergcement increased its presence as a supplier of bulk grey cement [over the last five years]. The effects of sales expansion are increasing pressure on the remaining competitors and activation of a large number of customers that are relevant for long-term market presence in Dalmatia."

(310) Fourth, the Parties are not distant competitors because of different product portfolios. Rather, CEMII and its varieties are by far the most important type of cement in Croatia and are offered by both Cemex Croatia and DDC.

(311) Fifth, the Commission recalls that the objections raised against the Parties were not premised on a contention that Cemex Croatia and DDC are each other’s closest competitors in the relevant markets. This Decision only finds that Cemex Croatia and DDC are close competitors in the relevant markets. This finding is not contradicted by the Parties’ claim that DDC and LafargeHolcim are Cemex’s “main competitors in Croatia” and is entirely aligned with the conclusions concerning DDC’s market shares in section 7.3 and the evidence on closeness of competition presented in sections 7.6.1 and 7.6.2.

61

ID225-8490, GL_C001_00019053.pdf, DDC 39063.

ID225-11284, GL_C001_00026370.pdf, DDC 26370.

ID225-10655, GL_C001_00024766.pdf, DDC 24766.

ID225-7288, GL_C001_00015533.PDF, DDC 15533.

ID1112-452, GL_C001_00011654 msg, DDC 11654.

ID1112-10730, GL_C001_00078124 msg.

ID1112-11970, GL_C001_00073083 msg.

ID1112-12335, GL_C001_00073496 msg.

ID1112-12328, GL_C001_00073487 msg.

ID1821, response of Strabag to Questionnaire to Customers (Q1), questions 27 and 16.

ID444, Form CO, paragraphs 207-208.

See responses to Questionnaire to Customers (Q1) dated 6, 7 and 15 September 2016, question 6, Questionnaire to Competitors (Q2) dated 6 September 2016, question 8, requests for information to customers dated 16 November 2016.

ID2410, HeidelbergCement's reply to the Statement of Objections, notably paragraphs 60, 68, 69.

62

7.6.3. Italcementi is an important competitive force in western Croatia

(312)Italcementi is an important competitive force in western Croatia where DDC's presence is more limited. In that area, Italcementi's grey cement, sold through its local distributor A.R.M.A.C.O.M., is a source of price competition as demonstrated by the following observation of DDC: “[content of internal documents].”

(313)Moreover, Italcementi and its local distributor, A.R.M.A.C.O.M., have been repeatedly targeted by various actions by Cemex Croatia seeking to impede imports of Italcementi’s cement into western Croatia, parts of which (Kvarner) are also covered by the relevant geographic markets. This indicates that Italcementi was perceived as a direct competitive threat by Cemex, after LafargeHolcim which is the most important cement supplier in western Croatia.

7.7. Constraints from competing suppliers of grey cement

(314)The competitors which transport grey cement over land to customers in the overlap area between the Parties' plants in Split and Kakanj are mainly Lafarge-Holcim, Asamer, Nexe, W&P, Colacem and Titan.

(315)For the reasons set out below in sections 7.7.1 to 7.7.5, the Commission concludes that the merged entity will face insufficient competitive constraints from competing land-based and sea-borne suppliers of grey cement after the Transaction. The Commission first sets out factors limiting the constraints exerted by both competing land-based and sea-borne suppliers of grey cement (section 7.7.1) before analysing factors that specifically limit the constraints exerted by competing land-based suppliers (sections 7.7.2 and 7.7.3) and sea-borne suppliers (section 7.7.4) of grey cement.

7.7.1. General factors limiting the constraints exerted by competing land-based and sea-borne suppliers of grey cement

(316)First, competing land-based and sea-borne suppliers of grey cement need to cover longer distance to reach the customers concerned by the Parties’ overlap, which entails higher transport costs.

All competing suppliers of grey cement in the overlap area between the Parties' plants in Split and Kakanj will be at a structural disadvantage vis-à-vis the merged entry because their production facilities are located at a considerable distance from many customers in the overlap area. This is confirmed by the additional driving distances to reach customers and the following statements:

(a) LafargeHolcim submits: "In relation to the Dalmatia region, Holcim Croatia believes that the higher prices in this region are due to the limited number of market players, due largely to its relative isolation and distance from most producers in Croatia. Transport distances and associated costs are therefore higher as well."

(b) According to Titan: "Titan imports Cement via Bosnia to Croatia, which entails high logistic costs due to the distance. … In the past, Titan tried to export bulk cement but gave it up due to logistic issues … Titan expects domestic transport costs to be 40-50% lower than costs for international transports from Serbia".

63

Second, and for the same reasons, land-based and sea-borne suppliers of grey cement will not be in a position to offer the security of supply as required by customers and as offered by the Parties.

(a) According to W&P: "If Cemex's cement prices in Dalmatia are higher we assume, that beside of the "premium" positioning of its cement brands also local availability from local producer, just-in-time delivery/timeliness of deliveries and reliability of deliveries, sales and technical support in Croatian language but also payment terms and conditions (acceptance of barter) could have an impact."

(b) According to Nexe: "Beyond transport costs […], we cannot guarantee frequent delivery of cement in the set timeframe for customers in Dalmatia. This is due to long distances from the place of loading to the unloading, closed roads due to bad weather, drivers who have to make regular breaks in driving the so far destinations and so on" ;

(c) According to Asamer: "Eventually warehouses for bag cement and terminals for bulk cement needed. The reliability of truck transport over such a long distance could be a risk."

(d) According to Titan, bulk cement customers have a "need for security of supply, which is achieved if there is proximity, i.e. presence with an integrated plant, or at least through the use of a distribution terminal or grinding plant".

(e) According to customer Dajaković: "In respect of imports from Turkey and Albania it considers that for such imports there are huge logistical problems, for example there are no cement terminals. Moreover, these [imports from Turkey and Albania] are not long-term relations, all is short-term, and they as RMX concrete producers cannot rely on such imports. As concrete producers the quality and possibility of constant and long-term supply of cement is extremely important to them."

(f) According to customer Dubac: "Considering that all [of Dubac's] three RMX plants together can store up to 500 t of cement, it cannot receive bigger quantities of cement and needs regular and reliable delivery. “Cemex and HeidelbergCement know that buyers cannot import cement from third [suppliers] that is that they have to buy cement from Split or Kakanj"

(319) Third, the incentives of competitors to supply grey cement are further curbed by possible actions by the Parties deterring entry or expansion (see further section 7.11).

(320) Fourth, importers of grey cement from certain countries may face to a certain extent a lower market acceptance and/or be required to offer greater price reductions:

(a) According to Cemex "products from non-EU countries can be perceived to be of a lower quality and have lower brand equity than EU-produced products".

(b) According to Titan "we have observed a preference of local customers towards Croatian products, comparing with our cement produced in Serbia" .

(c) According to customers in the relevant markets, they would consider purchasing imported cement from Albania, Serbia, Turkey, and Bosnia only if it were offered at discount of 10-35% on the price of grey cement produced in Croatia.

(321) DDC's imports appear to be less affected by such lower market acceptance in Croatia as set out by TCK in its annual report 2015: "In percentages, the share of exports in total sales TCK's increased by about 9% compared to the previous year. This suggests the stability presence and brand recognition on Croatian market."

(322) Fifth, importers of grey cement are at a disadvantage to win over new customers because of their reduced ability to engage into barter trading and to assess the creditworthiness of customers:

(a) According to W&P: "If Cemex's cement prices in Dalmatia are higher we assume, that beside of the "premium" positioning of its cement brands also local availability from local producer, just-in-time delivery/timeliness of deliveries and reliability of deliveries, sales and technical support in Croatian language but also payment terms and conditions (acceptance of barter) could have an impact."

(b) According to Asamer: "Domestic producers have an advantage for this kind of cooperation since they have wider business activity in Croatia than importers have (importers come only as suppliers and domestic producers are present also as customers of goods they need from local market). Some of competitors have operations in construction works (Nexe Našicecement) and they often use barter payment when sourcing other construction material they need. … At the moment, due to the fact that the market can be considered as customer's one, the brand, the image and geographic origin play minor role."

(c) According to Titan: "We do not consider that Cemex enjoys any exceptional premium in Dalmatia due to its brand name. As in other markets around the world established (integrated) producers, like Cemex, enjoy some added benefits due to their long standing presence in market, their traditional trading relationship with customers and the additional services they are able to provide."

64

(323) Sixth, the majority of customers that responded to the market investigation consider that in Dalmatia, imports from competing land-based and sea-borne suppliers of grey cement –other than those of DDC- do not exercise a significant competitive pressure on the merged entity after the Transaction.

(324) As set out below in section 7.7.2, the competitive constraints exercised by each competing land-based supplier of grey cement is further limited by a series of factors.

7.7.2. Constraints exerted by specific individual land-based suppliers of grey cement

7.7.2.1. LafargeHolcim

(325) The LafargeHolcim group is active in Croatia through its subsidiary Holcim (Hrvatska) d.o.o. Koromačno, which operates one integrated grey cement plant in Koromačno (western Adriatic Croatia) and two grey cement terminals, a sea terminal in Zadar (Dalmatia), and a land terminal in Jastrebarsko (Zagreb). LafargeHolcim is therefore mainly active in western Croatia in Istria&Kvarner (approximately [50-60]% sales share) and continental Croatia (around [10-20]%), mainly through deliveries from its Koromačno plant, and in Dalmatia (around [20-30]%), mainly through deliveries from its Zadar terminal.

(326) The Commission concludes that while LafargeHolcim will continue to compete with Cemex in the market, it will not sufficiently constrain the merged entity for the following reasons.

(327) First, the Koromačno plant is located at a considerable road distance from the Split plant of more than 400km i.e. more than 150km further than DDC's plant in Kakanj. The constraint that LafargeHolcim will exert from its Koromačno plant in the market around the Split plant will therefore be limited to certain areas in the north of that market, mainly Kvarner and Continental Croatia/the Zagreb region.

(328) Second, as its sea terminal in Zadar is capacity constrained, LafargeHolcim is unlikely to substantially increase its supply in the relevant markets if prices increase. The Zadar terminal is located at about 150km road distance from the Split plant and therefore closer to the centre of the relevant markets. The effective capacity of the terminal, which has a storage capacity of [0.8-1.5]kt, is [65-75]kt annually as measured by LafargeHolcim. In view of the fact that LafargeHolcim has sold [55-65]kt of cement on average in 2015-2016 from its Zadar terminal, the terminal does not allow for a significant expansion of sales to Dalmatia. Furthermore, regular variations in demand over the year, as described in detail in recitals (587)-(590) are likely to further limit LafargeHolcim's ability to constrain the merged entity through an expansion of sales from Zadar in times of peak demand when prices increases would be most pronounced.

(329) Third, contrary to the Parties' arguments, LafargeHolcim would not effectively increase the available capacity of its Zadar terminal by reallocating customers to be served directly from the Koromačno plant. Such reallocation would negatively impact LafargeHolcim's position around the Zadar terminal in terms of profitability as LafargeHolcim would achieve lower margins on the sales to the customers shifted from deliveries from the Zadar terminal to deliveries from the Koromačno plant (whether or not such impact is as "significant" as claimed by LafargeHolcim). Furthermore, LafargeHolcim would be able to offer the customers previously served from the closer Zadar terminal a lower reliability of services which is important to cement customers who value security of supply. LafargeHolcim has indicated that it would not consider reallocating sales and freeing up capacity at the Zadar terminal in that way: "it would effectively reduce margins by [a significant amount] (by increasing transport and logistics costs) as well as resulting in a slower response time (i.e. from order to delivery) for those customers who are closer to Zadar. From a commercial viewpoint it is therefore not a possible alternative".

(330) Fourth, contrary to the Parties' arguments, increasing the capacity of the Zadar terminal by increasing the number of shipments without investing into new storage capacity would not be feasible for LafargeHolcim. The ability of the Zadar terminal to increase shipments appears to be restricted by a number of factors:

(331) In the first place, maximising the capacity of the Zadar terminal would require shipments to be timely received at the moment when the stock at the terminal is as close to zero as possible, so that the maximum amount can be offloaded. Typical discharge time in Zadar takes approximately 10-12 hours, implying that if the shipment arrives when the silo is full, LafargeHolcim would have to sell the entirety of the capacity of the silo in the space of one working day in order to free the silo to receive new stock. According to LafargeHolcim: "Realistically this is not possible, and as a result the ship can often stay in the dock discharging the delivery from Koromačno for as long as it takes to free the silo (sometimes up to four days)".

(332) In the second place, it is difficult for LafargeHolcim to make more deliveries to Zadar with smaller ships. LafargeHolcim reports that: "it is difficult to arrange scheduling of smaller ships on a sufficiently regular basis to be able to serve Zadar enough to increase its effective capacity for two reasons – firstly, such smaller ships are difficult to contract and there are fewer available; and secondly, Holcim Croatia is often not the only entity scheduling delivery using the ship, so they often have to arrange the scheduling taking other entities into account. The difficulty of contracting smaller ships in the Adriatic Sea as well as the fact that those small ships available are contracted by different cement suppliers has also been confirmed by other market participants and will be summarized in the remedy section in recital (604) below.

(333) Fifth, LafargeHolcim has very limited commercial presence in Southern Dalmatia because its plant in Koromačno is located too far away -490km to Split- (in terms of logistics costs and security of supply) and because, in order to achieve better margins, LafargeHolcim uses the volumes available at its capacity constrained terminal in Zadar to supply customers closer to the terminal instead of customers located further away. That conclusion applies regardless of whether contribution margins are considered –which according to the Parties is the relevant benchmark- or total profits are considered. LafargeHolcim provided a confidential cost breakdown – comparing sales to the Zadar region and sales to the Dalmatia region – in support of those dynamics. It added that "even with an increase in prices to the region south of Split, the profit achievable on sales into the region would be [significantly] lower than into the Zadar region". The reasoning also applies in the same way if prices increased in the market after the Transaction. This was also confirmed by LafargeHolcim: "The outcome – i.e. reduced profitability on sales into the Dalmatia region regardless of price increases – remains equally valid when assessed either by reference to the level of contribution margin or the net income."

(334) Sixth, contrary to the Notifying Parties' submissions, LafargeHolcim would be unable to expand its presence in particular in Dalmatia through direct sales using road transport from its Koromačno plant. While LafargeHolcim supplies certain quantities of cement directly from its Koromačno plant to customers in Dalmatia, 60-70% of these sales concern bagged cement. The remaining bulk sales concerned special cement and exceptional deliveries made from the Koromačno plant "in order to fulfil Holcim Croatia's contract obligations despite the higher logistic costs incurred". LafargeHolcim “does not believe that the sale of additional volumes by road transport to Dalmatia from Koromačno is economically sustainable… Were Holcim Croatia to have a terminal in markets south of Zadar (for example in Split, Dubrovnik or even in Montenegro), it would be able to increase sales volumes in Dalmatia, but only through by [sic] seaborne transport”. However, no such terminals are currently available.

(335) Seventh, LafargeHolcim would not have the incentive to invest into new capacity in Dalmatia even in case of a price increase for cement, due to the need to construct a new silo for such capacity expansion and the barriers to such expansion in terms of construction and planning laws.

(336) Eighth, contrary to the Parties' claims, the Commission's conclusion that LafargeHolcim will not sufficiently constrain the merged entity is not called into question by the fact that LafargeHolcim has been supplying certain customers south of Split and in southern Bosnia.

(a) Concerning LafargeHolcim's sales south of Split, apart from two larger customers, Dajaković and C.M.A., LafargeHolcim does not have a broader commercial presence in that region. Dajaković has a specific commercial relationship with LafargeHolcim which is the reason for LafargeHolcim's sales to Dajaković. As for C.M.A., located in Imotski (180km north-west from Dubrovnik), it indicated that it was the southern-most customer of LafargeHolcim in Dalmatia.

66

(b) Concerning LafargeHolcim's sales in Bosnia, those are mainly limited to one customer. LafargeHolcim has indicated that "the majority of Holcim Croatia’s sales to Bosnia are made to [a long-standing customer located close to the border with Croatia who on-sells cement to RMX customers]". This is corroborated by [content of internal document]. Furthermore, LafargeHolcim does not incur any transport or logistic costs for serving this customer because the customer picks up the sales directly from the Koromačno plant. LafargeHolcim reports that it would have considerably lower margins if it transported the cement to customers itself. In that respect LafargeHolcim states that: "By comparison, [considering that sales made to the region south of Split are delivered, when deducting the sales and logistics costs incurred to make these sales the equivalent ex works price for sales from Koromačno would be significantly lower than in the case of sales to Bosnia]."

7.7.2.2. Nexe

(337) Nexe is a Croatian company which is active in selling cement to Croatian customers through its subsidiary Našicecement d.d. which operates one integrated cement plant in Našice in Slavonia. Nexe has also recently starting operating a small depot for bagged cement in Rijeka in western Croatia. Nexe is therefore mainly active in northern Croatia in Slavonia-Continental East (around 30-40% sales share) and in Continental Croatia (around 40-50%).

(338) The Commission concludes that Nexe will not sufficiently constrain the merged entity for the following reasons.

(339) First, Nexe's current commercial presence is limited to certain parts of the relevant markets, limiting the competitive pressure currently exerted by Nexe on the Parties. Nexe's plant is located in Našice in Slavonia at more than 600km road distance from Cemex' plant in Split (more than 350km further than DDC's plant in Kakanj is located from Split). While Nexe is able to offer grey cement at competitive prices in Continental Croatia and Slavonia-Continental East, its high transport costs result in limited sales quantities in Istria - Kvarner and Dalmatia.

(340) Second, Nexe uses external transport providers and has to rely on the availability of return cargo arrangements to lower transportation costs to reach customers in Dalmatia. However, the availability of such return cargo arrangements is limited for Nexe: "transport providers do not have two way transport tours between Slavonia and Dalmatia, which results in higher transport costs. That is corroborated by Cemex itself, which stated that: “As there is very little industry and production in the south of Croatia, many trucks that deliver goods and products into the south of Croatia return north empty.”

(341) Third, Nexe would have to cross long distances to reach customers in more western and southern parts of Croatia since Nexe currently does not supply cement to such customers by transiting through Bosnia-Herzegovina and such transit is not feasible, as set already out in the context of the geographic market definition in recitals (200) to (207). Nexe's ability to supply cement in Croatia by crossing Bosnia-Herzegovina is limited specifically for the following main reasons:

(342) In the first place, deliveries through Bosnia-Herzegovina appear to take considerably longer than deliveries through Croatia. Publicly available route calculation applications indicate that, for heavy trucks, driving times from Našice to Split through Croatia are comparable to those for transit through Bosnia-Herzegovina excluding waiting times at border crossings. However, transiting through Bosnia-Herzegovina would involve two border crossings and "extend the time for delivery (shortened procedure could not be applied) which considerably raises the transport costs and delivery time."

(343) In the second place, Croatian hauliers would require Bosnian work permits and there are barriers to using Bosnian hauliers as transit through Bosnia-Herzegovina, as a non-EU country, is restricted by cabotage rules. Articles 94 to 96 of the Croatian “Road Transport Act” prohibit foreign carriers from performing internal transport (cabotage) in the territory of Croatia. Bosnian hauliers cannot therefore perform transport of cement from Slavonia to Dalmatia (or vice-versa) through Bosnia-Herzegovina. This is confirmed by Keš d.o.o. ("Keš"), a construction material trader which, amongst others, distributes Nexe cement in Bosnia-Herzegovina.

(344) In the third place, contrary to HeidelbergCement's claims, it is not economically viable to import cement into Bosnia-Herzegovina and subsequently export it to Croatia to circumvent the cabotage restrictions.

(a) According to HeidelbergCement, such a practice would result in "additional costs for freight forwarding and customs and a custom duty fee of […]% of the goods' purchase price".

(b) According to Cemex Croatia, "there are no customs fees but forwarding an agent fee can affect up to […] EUR per delivery ([…] EUR for export in Bosnia and […] EUR for import in Croatia)" even if for regular imports fee can be significantly lower.

(345) In the fourth place, rail transit through Bosnia-Herzegovina is equally not feasible for Nexe due to the complexity of organising rail transport with the different rail operators.

(346) In the fifth place, while Nexe sells cement to distributors in the south of Bosnia-Herzegovina (most of the transport is organised by customers themselves), it does not serve Dalmatia or any other part of Croatia by transit through Bosnia-Herzegovina.

(347) In the sixth place, Cemex’ deliveries of cement from Split to Slavonia do not transit through Bosnia-Herzegovina although Cemex supplies a number of Bosnian customers […].

(348) Fourth, Nexe is unable to guarantee customers in Dalmatia frequent deliveries of cement within requested deadlines due to the distance between its plant and the customers as well as other factors such as occasional closures of transport routes due to weather conditions or regulated breaks for truck drivers.

(349) Fifth, only 3 customers who responded to the market investigation and are located in Dalmatia mentioned Nexe as a potential supplier of grey cement.

(350) Sixth, Nexe would have significant difficulties to expand its sales to those areas in Croatia where it is today hardly present, i.e. western and southern Croatia, and in particular in Dalmatia:

(351) In the first place, Nexe has been in financial difficulties in recent years, and entered into pre-bankruptcy settlements in 2014 and 2016. That may affect its ability and/or timing to finance investments into expansion, including investment in logistics facilities such as terminals or rail connections.

(352) In the second place, Nexe could not expand its direct sales through road deliveries due to the abovementioned high costs of transports and security of supply issues (see section 7.7.1) and accordingly "does not expect high increase of cement sales through road transport of either bulk or bagged cement." Nexe claims that "[w]ithout a silo terminal its stable presence in Dalmatia, expansion of activities and supply of bigger quantities of bulk cement is not feasible."

(353) In the third place, Nexe could not expand its sales in the market through the use of an existing rail connection and is unlikely to invest in a new rail connection to reach customers further south of its current sales area.

(a) Currently, Nexe delivers limited quantities of bagged grey cement by rail to a warehouse in Rijeka (Kvarner) which is located on the rim of the 250km circular catchment area around Split. That warehouse is only suitable for sales of bagged grey cement and not for bulk grey cement. Moreover, due to the topography and availability of highways, using the warehouse for sales to customers further south would mean that the trucks would travel back east for 90km before going south with the ensuing additional costs.

321

(b) Nexe itself states that it is difficult to estimate the time needed to complete a new rail connection in view both of the administrative/permitting procedures and of financing. The completion of the investment to construct and connect the plant to the public rail for the Rijeka warehouse took 10 years, was financially very demanding and involved long re-negotiations with the Croatian Railways and a protracted administrative process regarding location and building permits.

(354) Sixth, contrary to HeidelbergCement's claims, the Commission's conclusion is not called into question by the fact that Nexe supplies grey cement to Keš, a customer located in southern Bosnia. Keš does not consider expanding to Dalmatia and does not expect such expansion by Nexe. This is because grey cement supplied by Nexe from its plant in Našice is less price-competitive close to the southern border between Bosnia and Croatia due to high transport costs, allowing only for supply distances of up to 350km.

7.7.2.3. Titan

(355) Titan is a Greek building materials company which operates integrated cement plants in Serbia and Albania among other locations. Titan currently supplies cement from its Serbian plant in Kosjerić to Croatia and focusses its sales on Dalmatia (around 5-10% sales share) and Slavonia-Continental East (around 0-5%).

(356) The Commission concludes that Titan will not sufficiently constrain the merged entity for the following reasons.

(357) First, Titan makes limited sales of exclusively bagged grey cement – accounting for up to [5-10]% share in the relevant markets – in those areas that it can access by road from its plant located in Serbia (Kosjerić) to the extent that transport costs allow competitive pricing. Dalmatia is the nearest point of entry into Croatia, of which Titan can serve only a limited part in the south.

(358) Second, Titan incurs high logistic costs when supplying its bagged grey cement in Croatia due to the distance. Titan's plant in Serbia is located more than 450km away from the Split area (more than 200km further than DDC's plant in Kakanj is located from Split). In order to ensure competitive transport costs, Titan's distributor in Croatia must use two-way transport where possible, which increases further the complexity of logistics: "Cement from Kosjeric to Croatia is transported by 24-tons trucks, with two-way transport to the degree possible in order to save transport costs".

(359) Third, due to the lack of infrastructure, in particular the absence of a highway connection between Serbia and Croatia, imports to Croatia are difficult for Titan and entail average estimated traveling time of 16 hours.

(360) Fourth, Titan cannot import bulk grey cement into Croatia as it does not have a solution for bulk cement logistics for that distance which puts it at a competitive disadvantage in comparison to other suppliers: "The disadvantage of not selling bulk cement is that sales of bulk are higher in volume as they are usually targeted to major construction projects." Titan can therefore address only up to roughly […]% of demand while not being an option at all for the majority of customers.

(361) Fifth, in the past, Titan unsuccessfully tried to export bulk grey cement to Croatia. It gave it up due to logistics issues. Most notably, customers primarily purchasing bulk grey cement value more security of supply relative to customers purchasing mostly bagged cement.

(362) Sixth, while Titan has considered the possibility of exporting bulk cement by rail to Croatia, it considers this solution unfeasible because the distance that its bulk cement would have to travel by rail would be much longer in comparison to delivery by truck.

(363) Seventh, Titan lacks brand reputation and certain customers are said to be unwilling to source from Titan due to the Serbian origin of the cement and others report poor quality.

(a) According to Titan: "there is tacit perseverant aversion to Serbian imports into Croatia for reasons linked to recent history" and "We have observed a preference of local customers towards local Croatian products, when comparing with our cement produced in Serbia".

(b) According to DDC's internal commercial discussions:

345 ID225-1409, M.7878 - QP 2 Section B (E-mails request) - USB Stick GL_C001_00002573.xlsx.

346 Ibid.

347 [Content of internal documents].

(364) Eight, Titan currently sees no scope for expansion of its imports into Croatia:

(a) "Overall for Titan it would be hard to ensure viability of expanded sales in Dalmatia at the moment. Titan has no plans to expand its sales in Dalmatia. Titan owns a relatively small plant that is running currently at a capacity of approx. 50%. Taking into consideration the expected growth in Serbia and Montenegro in the next 5 years due to large infrastructure projects, Titan estimates that there will not be sufficient idle capacity to commit sustainably to increased sales into Croatia. Regarding the current pricing environment and logistics Titan does not see a lot of potential in the Croatian market. … Looking at the overcapacities, the market situation and the investments required, Titan has no plans to open a grinding plant or a cement terminal in Dalmatia."[…]

(b) "If we were to decide to import bulk cement to Dalmatia, we would probably also decide to invest in a distribution terminal or a grinding plant. However, in view of the conditions currently applicable in the market, where there is a surplus in the supply volumes, such a capital investment in a terminal or a grinding plant does not seem economically justified."

(c) "Furthermore, due to Titan's existing penetration in the bagged cement segment there appears to be limited scope of growth even in case of a price increase."

(d) "Even if cement prices increased in Dalmatia, Titan is sceptical about the possibility of a large long term expansion – pushing volumes could erode the price and the long term sustainability of such imports for an importer with high transport costs."

(365) Ninth, out of the 43 customers replying to the Commission's investigation, only six customers referred to Titan as a potential supplier of bagged cement. These customers are all located in Dalmatia, which is, as mentioned in recital (357), the nearest point of entry into Croatia for Titan's production locations. None of the customers refers to potential supplies from Titan's production in Albania, but refer only to its plant in Kosjerić (Serbia). A customer indicated in this regard, "Imports from Serbia is not a solution, although Titan offered better conditions, however, it cannot rely in the supply of cement on a supplier which is not 100% reliable considering the distance and crossing of 2 to 3 borders."

345 ID225-1409, M.7878 - QP 2 Section B (E-mails request) - USB Stick GL_C001_00002573.xlsx.

346 Ibid.

347 [Content of internal documents].

7.7.2.4. Asamer

(366) Asamer is an Austrian building materials company which is present in Croatia through cement imports from its integrated cement plant in Lukavac near Tuzla in northern Bosnia-Herzegovina. Asamer therefore focuses its sales in Croatia on Slavonia-Continental East (around 5-10% sales share) and Continental Croatia (around 0-5%).

(367) The Commission concludes that Asamer will not sufficiently constrain the merged entity for the following reasons.

(368) First, Asamer is not active in Dalmatia and makes limited sales of bagged and bulk grey cement – accounting for up to [5-10]% share in the relevant markets – only in the areas of Slavonia and Continental Croatia that it can access from its plant in Lukavac, Bosnia-Herzegovina. The plant is located 350km away from the Split area (more than 100km further than DDC's plant in Kakanj is located from Split).

(369) Second, in Croatia, Asamer focusses its activities in the region of Zagreb where demand is stable all year long and to where it is less burdensome to supply: “At the moment we do not sale [sic] in Dalmatia but in Zagreb due to the fact that Zagreb represents almost 50% of total market in Croatia and there is activity present all over the year: It is correct that the distance is similar but time vise[sic]; Zagreb is much closer (5-6- hours compared to 8-10 hours to Dalmatia). Also, the road to Dalmatia is in very bad condition (only 70 kilometres is highway) and during winter months (due to weather conditions – heavy snow as usual matter) we cannot insure delivery at any case.”

(370) Third, Asamer is consistently not considered by the Parties as an aggressive competitor:

(a) According to the DDC Flash Reports for Bosnia-Herzegovina for the period October 2014-December 2015 [content of internal documents].

(b) According to the Cemex "Balkan Strategy – Threats and Opportunities Draft" (CX172) from 2014 Asamer was [content of internal documents].

(c) According to another Cemex document from March 2013 Asamer focuses [content of internal documents].

(371) Fourth, Asamer focusses its activities on the Bosnian region immediately adjacent to its production plant where the Parties refrain from selling significant volumes (despite the closeness of DDC's plant to that region). This indicates that Asamer and the Parties do not compete aggressively even in regions closer to Asamer's own plant in Bosnia-Herzegovina:

(a) Cemex's direct and indirect sales in Bosnia-Herzegovina to customers within 150 road km around Asamer's plant in Lukavac amounted to […]kt in 2015, while the corresponding sales of DDC in the same area amounted to […]kt.

(b) The Parties' sales to customers within 100 road km around Asamer's plant in Lukavac in 2015 amounted to […]kt and […]kt, respectively for Cemex and

ID2102, reply to questionnaire of 10 August 2016.

ID2242, reply to request for information of 09 November 2016, Q19.

ID405-59, M.7878_Form CO_Annex 5.4.III.-RFI 20160608_QP1_Q13b.11_DDC Kakanj_Bosnia Flash Reports 2014.PDF.

ID437-28, "Balkan Strategy – Threats and Opportunities Draft" (CX172).

ID437-39, M.7878_Annex_RFI 20160711_QP7_Sales Mar 2013 (CX 164).PDF.

DDC. The area within 100 road km around Asamer's plant in Lukavac essentially represents what DDC [content of internal documents and strategic information].

(c) The 100 road km around Asamer's plant in Lukavac is a high consumption area. DDC estimates that the total consumption of cement in [strategic information] in Bosnia-Herzegovina is […]kt in 2015, with Asamer supplying […]% of the cement, while DDC and Cemex having […]% and […]% market shares, respectively.

(d) As it will be explained in section 7.11, Asamer’s presence in Bosnia-Herzegovina and Croatia suggests a reluctance to enter into the traditional markets of Cemex and DDC in southern Bosnia-Herzegovina.

(372) Fifth, Asamer's sales to Southern Croatia would entail higher travel distances compared to those for Zagreb, and the ensuing difficulties in ensuring the continuity of supplies throughout the year.

(a) due to the bad road conditions in Bosnia-Herzegovina, it would take Asamer three hours more than DDC to reach customers in Dalmatia.

(b) while the ongoing construction of a highway through Bosnia-Herzegovina to South Croatia will facilitate the transport in the future, it will take another 3 to 5 years until the project is finished.

(373) Sixth, Asamer's ability to compete in the Istria and Kvarner region and in Dalmatia is limited by transport costs of between 17-23EUR/t (based on return cargo). By contrast, only a few customers of Asamer from Southern Bosnia-Herzegovina buy cement ex work Asamer's plant in Lukavac and transport the cement themselves. According to information provided by Keš, located in southern Bosnia, Keš does not purchase cement from Asamer and does not expect Asamer to expand to Dalmatia.

(374) Seventh, in order for Asamer to offer the requisite security of supplies to customers in Dalmatia – including for potential additional sales in Dalmatia –,“Eventually warehouses for bag cement and terminals for bulk cement needed. As reliability of truck transport over such a long distance could be a risk.” Furthermore, to expand, “Fabrika cementa Lukavac would have to build up sales organisations for Dalmatia market, make detailed market analysis, contract quantities with customers and organise logistics (for permanent supply terminals/warehouses to be considered)”. Asamer has, however, not taken any steps in that direction: “due to risks in logistics, high cost and additional high fixed costs we don’t consider it at the moment.”

ID437-103, Annex 5.4-I-(11) of the Form CO – HC Market Study Bosnia and Western Balkan.

ID437-103, Annex 5.4-I-(11) of the Form CO – HC Market Study Bosnia and Western Balkan.

ID2242, reply to request for information of 09 November 2016.

ID2111, minutes of a telephone interview with Asamer of 30 November 2016.

ID2111, minutes of a telephone interview with Asamer of 30 November 2016.

ID1350, reply to a request for information of 21 November 2016, Q1(c).

ID2099, reply to a request for information of 09 November 2016.

(375) Eighth, while in 2013 Asamer exported cement to Libya via the Port of Ploče, situated in Dalmatia, that was a one-off in response to the specific competitive situation faced by Asamer in Libya at the time: “In addition, export to Libya was not motivated only by commercial terms and profits but by the fact that owners of Fabrika cementa Lukavac were at that time owners of cement plants in Libya also. Due to Libya crisis, cement plants in Libya were not able to produce and fulfil local market demand and our common owners tried to cover missing quantities with Bosnian export. The price was adjusted to be competitive on Libyan market…. Sale to Libya was short term sale in order to help our Group Libya operations (cement production). It was a „ad-hoc“solution (project), not long term continuous cooperation. The goal was to deliver up to 30.000 tons (6 ships) but due to Libya crisis escalation, export was stopped.”

(376) Ninth, only two Croatian customers that replied to the market investigation, – one of which is located in Dalmatia – mentioned Asamer as a potential supplier of grey cement.

7.7.2.5. Wietersdorfer & Peggauer ("W&P")

(377) W&P is an Austrian building materials company which supplies cement to Croatia mainly from its integrated cement plant located in Anhovo in western Slovenia close to the Italian border. W&P therefore focuses its sales in Croatia on Continental Croatia (around 5-10% sales share).

(378) The Commission concludes that W&P will not sufficiently constrain the merged entity for the following reasons.

(379) First, W&P makes limited sales of grey cement only in parts of the relevant markets (Central Croatia and Slavonia-Continental East that overlap only partially with the relevant markets around Split) that it can access from its production facilities in Italy (San Vito al Tagliamento) and Slovenia (Anhovo), both being more than 500km road distance away from the Split area (more than 250km further than DDC's plant in Kakanj is located from Split). W&P has not made any sales in Dalmatia for the past 10 years.

(380) Second, in order for W&P to offer the requisite security of supplies to customers in Croatia a "sales infrastructure in terms of a network of distributors would be needed".

(381) Third, in order to start supplying in Dalmatia W&P would need to overcome the (i) lower creditworthiness of customers in Croatia, (ii) prevalence of non-monetary/in-kind compensations ("barter") in Croatia, (iii) lack of substantive market knowledge regarding Croatia, and (iv) unforeseen border closings. W&P concludes "i n these conditions, it may not be economically sensible to transport cement for 500km".

ID1350, reply to a request for information of 21 November 2016, Q 4(a).

See response to Questionnaire to Customers (Q1) of 6 7 and 15 September 2016, Question 9.2. and minutes of a telephone interview with a customer/trader 8 July 2016 (ID965).

Based on the document ID437-103, M.7878_Form CO_Annex 5.4.-I.-(11)_HC Market Study Bosnia and Western Balkan.PDF.

ID2166, reply to request for information of 09 November 2016, Q21.

ID2166, reply to request for information of 09 November 2016, Q23.

ID2167, minutes of a telephone interview with W&P of 30 November 2016.

7.7.2.6. Cementizillo

(382) The Commission concludes that Cementizillo will not sufficiently constrain the merged entity for the following reasons.

(383) First, Cementizillo itself is not active in Croatia its grey cement being sold in Croatia through a distributor, BFB. Cementizillo itself lacks the language and market knowledge required to operate in Croatia.

(384) Second, BFB sells limited quantities of Cementizillo grey cement in Croatia (7-7.5kt). Moreover, those quantities are decreasing.

(385) Third, BFB has a limited geographic coverage in Croatia, up to around 300km from Cementizillo's plant in Fanna (Friuli-Venezia Giulia), preventing it from make sales further south than Zadar.

(386) Fourth, BFB has a limited product portfolio. It only sells bagged grey cement, and focuses on two types of specialty grey cement, namely Pozzolanic cement CEM IV and fast setting CEM II 32,5 which command higher prices and support the transport costs.

(387) Fifth, BFB explained that, in the case of a price increase from the Transaction, it would be unlikely to react aggressively but would rather follow and increase its own prices.

7.7.2.7. Grigolin

(388) The Commission concludes that Grigolin will not sufficiently constrain the merged entity for the following reasons.

(389) First, Grigolin imports very limited grey cement volumes into Croatia from its grinding station in Susegana/Treviso (approximately […]kt in 2015) and has only two retailer customers exclusively in Istria which falls outside of the relevant markets (see for instance Figure 6 and Figure 14). The Parties estimate its market share to be about [0-5]% in the region of Istria-Kvarner (see Table 9).

(390) Second, Grigolin does not have its own established logistics solution for deliveries into Croatia as its customers in Croatia purchase its grey cement ex-works.

(391) Third, Grigolin does not consider the Croatian market as a business opportunity and would not increase its exports to Croatia even in case of a 5-10% price increase there.

7.7.3. Competitive constraints exerted by other potential land-based importers

(392) The Commission concludes that no other land-based importers will be able to sufficiently constrain the merged entity in the event that, after the Transaction, the latter would engage in price increases.

(393) First, Cimsa, located in Turkey, has never been active in Dalmatia and would not start selling in that region in the event that, after the Transaction, the merged entity would engage in a price increases as it "would not be sufficient to support Cimsa's sales in Dalmatia".

(394) Second, Limak, located in Turkey, has never been active in Dalmatia and would not start selling in that region in the event that, after the Transaction, the merged entity would engage in a price increases, due to the road distance between its production facilities and Croatia as well as between its production facilities and the nearest seaport.

7.7.4. Factors limiting the constraints exerted by competing sea-borne suppliers of grey cement

(395) First, transport costs and security of supply considerations will put competing sea-borne suppliers of grey cement at a particular competitive disadvantage compared to the merged entity (see recitals (316)-(318)), due to the longer distance the former have to cross to reach customers in Croatia.

(396) Second, while no terminals are needed to import bagged grey cement, such imports would be more limited in volume (due to handling of pallets/large bags) and would not be a suitable offering to the most important customer group – ready-mix concrete producers, which source bulk grey cement:

(a) According to a trader: “Small imports of bagged cement by boats cannot represent significant competition. When imported by boat bagged cement suffers damage and achieves lower price. Cement in big bags is fragmented in the port and transferred into cisterns. This procedure is only a "first aid" as it causes dust and high handling costs. This process could be acceptable for smaller quantities (500 t), however, for bigger quantities this proceeding is not acceptable.”

(b) According to a supplier: “We do not believe that bulk cement could be sold in big bags, due to the particularly high costs of handling and the complex unloading process for bulk silo loading at a customer site, which decentivise big bag imports into Croatia.”

(c) According to the Parties' own data big bag transports are [amount] used. Between 2012 and 2016, HeidelbergCement's entire EEA operations used big bags only for transport [relevant countries].The volumes concerned amounted to less than [amount] over the span of [amount] for the [amount].

(d) In the past 5 years there have been no substantial attempts of importers to continuously import bagged grey cement in the port of Ploče and Split.

(397) Third, while large scale distribution (and potentially bagging) of bulk grey cement is only possible through appropriate port infrastructure, in particular access to sea terminals, all existing sea terminals in Croatia are currently owned or operated by Cemex or LafargeHolcim.

(398) Regarding the need to have access to appropriate port infrastructure, in particular terminals:

(a) According to one trader: "In order to achieve international seaborne imports there has to be a silo in the port. For example Turkish producer has silo in Trieste. Holcim has silo in Zadar.”

(b) According to one supplier considering access to a sea terminal necessary for significant and regular seaborne imports of grey cement: “Unloading of bulk cement without a terminal is a complex and expensive process. The only option for unloading bulk cement without a terminal may be to use bulk carriers as floating silo until the carrier is emptied. That option is probably too expensive for suppliers due to the waiting time of the vessel which is inherent in the practice.”

(c) According to a HeidelbergCement internal document the Transaction will [content of internal documents] in the broader Western Balkans region based, amongst other, on the argument that [content of internal documents].

(d) According to a number of customers deficient port infrastructure is a barrier to significant seaborne imports of grey cement.

(399) Regarding the fact that all existing sea terminals for cement for the import of bulk grey cement are currently owned or operated by Cemex and LafargeHolcim:

(a) Cemex Croatia currently owns or operates seas terminals in Bakar/Western Croatia, and in Montenegro, close to the border with Croatia. Cemex Croatia also operates a river terminal in Metković/Dalmatia.

(b) LafargeHolcim currently owns a sea terminal in Zadar, Dalmatia.

(c) While the port of Split provides facilities for the import of bagged, but not bulk, grey cement, there have been no attempts to import bagged grey cement via the Port of Split in the last five years.

(d) It is also unlikely that the Port of Split will make the necessary investments in the special facilities needed to handle bulk grey cement, for lack of budget.

(e) There are no cement silos in Croatia for large ships enabling imports of bulk grey cement. The port of Ploče owns coastal silos for other purposes which are not suitable for cement. An existing alumina silo could be converted, but that would require significant investment and would raise environmental and also legal issues, as the silo is co-owned by the alumina-operator. Moreover, to use the plot that is currently used for coal for a cement silo would require strong economic justifications.

(f) Cement silos in the port of Dubrovnik were dismantled in 2015 and there are currently no silos dedicated for the import of cement. It is also unlikely that a new silo will be constructed as the port is focussed on tourism.

(400) Fourth, there have been no sea-borne imports of grey cement in Croatia in the last five years. Moreover, prior to that, sea-borne imports of grey cement were only sporadic and limited in volume: between 2009 and 2010 Turkish producer DENIZLI sourced approximately 50,000 tons of bagged cement using the port in Split. However, such imports were limited to ad hoc deliveries by an alliance of Dalmatian ready-mix concrete operators, and failed to establish a sustainable sales channel to Croatia.

(401) Fifth, Cemex Croatia's lease on the Metković terminal will expire, [information on Cemex's lease agreement]. Indeed, internal documents of the Parties from the period prior to the Transaction suggest that absent the Transaction the Parties may prolong the lease to prevent entry by other suppliers of grey cement.

(402) Moreover, although the Notifying Parties have in the meantime offered to terminate the lease of the Metković terminal, that offer has been made only as part of the commitments offered with a view to rendering the Transaction compatible with the internal market and is thus conditional on the Commission declaring the Transaction compatible with the internal market. That offer will therefore be assessed only in section 9 of this Decision. For the same reasons, the signing of a lease agreement by Asamer and Port of Ploče on 13 March 2017 and the signing of a termination agreement between Cemex and Port of Ploče of 16 March 2017, which are likewise conditional on the clearance of the Transaction by the Commission, will be assessed only in section 9 of this Decision.

(403) Sixth, sea-borne suppliers of grey cement have neither the incentive nor the ability to invest in a new sea terminal, either pre- or post-Transaction: “In respect of the construction of terminals in the ports, considers that it could be profitable only if it is likely that it will be possible to sell certain bigger quantities of cement (150-200 kt per year). Dalmatia is too small market for that to be profitable.” Moreover, the construction of a new sea terminal would require long lead times: “It is technically possible, but it would be a quite complex and long process which would require significant investment of time and (to a lesser extent) finances.”

7.7.5. Cumulative constraints from competitors

(404) In its reply to the Statement of Objections, HeidelbergCement argued that the Commission’s assessment failed to analyse the competitive pressure exercised cumulatively by all existing competitors with spare capacity, and instead merely analysed every competitor in isolation.

(405) The Commission concludes that the merged entity would not be sufficiently constrained by competitors after the Transaction, even taken cumulatively based on: (i) the specific assessment of actual or potential competitors’ potential to constrain the merged entity in the preceding sections 7.7.1 to 7.7.4; and (ii) the analysis in section 7.8 which shows that: the merged entity will face insufficient competitive pressure from spare capacity of competing suppliers of grey cement; and competing suppliers of grey cement will have insufficient incentive to expand sales to the relevant markets sufficiently to eliminate the risk of post-Transaction price increases in the relevant markets, including Dalmatia. The conclusions of that section apply equally to a cumulative response of these competitors.

7.8. Spare capacity and incentives to expand of competing suppliers of grey cement

(406) The Parties claim that spare capacity of grey cement will exercise an important competitive constraint on the merged entity with free capacities being sufficient to prevent price increases or output reductions following the Transaction. Each of the geographically closest competitors to the Parties (LafargeHolcim, Asamer, Nexe and Titan) as well as overseas suppliers located on the Adriatic and Mediterranean Sea would have sufficient spare capacity to replace DDC's sales in the relevant markets. Moreover total free capacity of rivals would be largely sufficient to cover the entirety of grey cement consumption in Dalmatia and other areas of overlap.

(407) Moreover, in the event of a price increase by the merged entity following the Transaction, competitors would have the incentive to serve the relevant markets, including Dalmatia, sufficiently to eliminate the risk of such a price increase. To support this claim, the Notifying Parties argue that competitors would be able to achieve positive contribution margins in serving virtually all of DDC's customers in Dalmatia, even if the competitors applied a 10% discount to DDC's current prices.

indicates "the Parties wish to terminate the Existing Lease Agreement, provided that the European Commission accepts the Commitments and clears the Transaction on this basis".

(408) For the reasons set out below in recitals (409) to (419), the Commission concludes that post-Transaction: (i) the merged entity will face insufficient competitive pressure from spare capacity of competing suppliers of grey cement; and (ii) competing suppliers of grey cement will have insufficient incentive to expand sales to the relevant markets sufficiently to eliminate the risk of post-Transaction price increases in the relevant markets, including Dalmatia.

(409) First, any spare capacity in the relevant markets or in the wider region is not sufficiently available to be used (cost-effectively) as it is located in plants that are considerably further away from Cemex's Split plant than DDC's Kakanj plant. This will result in permanent competitive disadvantages for competing suppliers of grey cement due to higher transport costs, issues of security of supply and other disadvantages (see section 7.7.1) and the merged entity's potential to curb expansion or entry (section 7.11).

(410) Second, the incentives of competing suppliers of grey cement to expand are likely to be insufficient to defeat price increases after the Transaction. Instead, it is likely that competitors will respond by increasing their prices in turn to benefit from higher volumes at higher prices. This is for the following reasons.

(411) In the first place, the absence of other competing suppliers of grey cement in Dalmatia despite the positive contribution margins at pre-merger prices indicates that those suppliers currently do not have the incentives to expand and supply grey cement in the relevant markets. According to the Notifying Parties' logic, the prospect of gaining positive margins should give competing suppliers of grey cement the incentive to expand their sales in the relevant markets already today. However, as indicated in Table 9 and Table 10, only LafargeHolcim and the Parties currently make significant sales in Dalmatia.

(412) In the second place, Asamer, Nexe and Titan currently have no or only a limited presence in the Mostar region of Bosnia-Herzegovina identified by DDC as region 5, Mostar region (see Figure 22).

Figure 22: Regions of Bosnia-Herzegovina as used by DDC; the regions discussed here are region 5, which is adjacent to southern Croatia/Dalmatia, and region 4

[…]

(413) According to DDC's internal calculations for 2015, the estimated market shares in region 5 in Bosnia-Herzegovina were [60-70]% for Cemex, [30-40]% for HeidelbergCement/DDC and [5-10]% for Nexe while [market share estimates] was assigned to Asamer, Titan or any other competitor.

(414) Asamer, Nexe and Titan have limited or no market share in the Mostar region despite margin levels being significantly higher than in Dalmatia:

(a) Average delivered prices of CEM II bulk grey cement (the cement type mostly sold in the area) in region 5 of Bosnia-Herzegovina are approximately […] EUR/t higher than prices in Dalmatia (approximately […] EUR/t in region 5 of Bosnia-Herzegovina, and […] EUR/t in Dalmatia). This is confirmed by DDC's internal documents and by HeidelbergCement's reply to the Statement of Objections: "[content of internal documents and pricing information]".

(b) Delivering grey cement in region 5 of Bosnia-Herzegovina is cheaper than delivering grey cement in Dalmatia for Asamer (located in northern Bosnia-Herzegovina), Nexe (Slavonia/Croatia) and Titan (Serbia). The transport cost difference is between […] EUR/t (based on the Notifying Parties' estimates of the additional transport costs) and […] EUR/t (based on DDC's own transport costs) for those competing grey cement suppliers.

(c) The margins of Asamer, Nexe and Titan in region 5 of Bosnia-Herzegovina are thus on average […] to […] EUR/t higher than the corresponding margins in Dalmatia, which is equivalent to about […]%-[…]% of DDC's prices.

(415) In the third place, similar reasoning applies to the region of Bosnia-Herzegovina around Kakanj that DDC identifies [strategic information], where Asamer, Nexe and Titan have only a limited presence. According to DDC's internal calculations for 2015, the estimated market shares in [strategic information] Bosnia-Herzegovina were [5-10]% for Cemex, [70-80]% for HeidelbergCement/DDC, [5-10]% for Asamer, and [5-10]% for Nexe while no market share was assigned to Titan or any other competitor.

(416) Asamer, Nexe and Titan have limited or no market share in the Sarajevo region despite: (i) [content of internal documents]; and (ii) transportation costs for Asamer, Nexe and Titan being lower because these firms' plants are significantly closer to region 4 of Bosnia-Herzegovina than to Dalmatia.

(417) Moreover, DDC has […] contribution margins in the Sarajevo region in the range of […]%, which further indicates that these competing suppliers do not seek to expand their sales in that region to such an extent that DDC's margins would be under significant competitive pressure.

(418) Third, in light of the presence of positive margin opportunities in the relevant markets in combination with differentiating features of those markets (variations in transport costs and security of supply due to geographic location as well as other factors such as quality and reputation), the elimination of competition between Cemex and DDC is unlikely to be offset by the presence of more distant suppliers of grey cement that are at a competitive disadvantage. Rather, competing suppliers of grey cement will have an incentive to expand their output only to a limited extent in order to benefit from increased demand at higher prices.

(419) Fourth, the conclusion that competing suppliers of grey cement will have insufficient incentives to expand sales to the relevant markets sufficiently to eliminate the risk of post-Transaction price increases in the relevant markets, including Dalmatia is not affected by the existence of customer specific price negotiations:

(a) The fact that the Parties' maximum effects analysis – which implicitly models bidding by grey cement suppliers that are differentiated by their location for individual customers in the relevant markets through a fully transparent process – predicts price increases illustrates that post-merger expansions by the merged entity's rivals are not sufficient to eliminate post-Transaction price increases in the context of a transparent bidding process.

(b) Even when there is imperfect transparency regarding the identity of competing bidders, their costs, bids and/or customer preferences competing grey cement suppliers will have the incentive to bid less aggressively for customers following a price increase by the merged entity and will therefore not defeat a price increase. This is because with imperfect transparency, firms will face a trade-off between the level of the price they offer and their expected volume similar to that in standard differentiated markets, with the consequence that the reasoning in recital (418) also applies to this case.

7.9. Likelihood of entry by other suppliers of grey cement

(420) For the reasons set out in recitals (421) to (430), the Commission concludes that also entry by other suppliers of grey cement into the relevant markets cannot be considered a sufficient competitive constraint on the merging Parties since such entry would not be likely, timely and sufficient to deter or defeat any potential anti-competitive effects of the merger.

(421) First, regarding the intention of ArcelorMittal to open a slag grinding station within its existing steel plant in Zenica, Bosnia-Herzegovina, that project is unlikely to grow within a sufficiently short period of time into a viable competitive force that could compete effectively with the merged entity on a lasting basis.

(422) Furthermore, the potential of a slag grinding station producing blast furnace slag to constraint the Parties is limited as blast furnace slag can substitute cement only to a limited extent. As explained in footnote 103, ground granulated blast furnace slag is an alternative cementitious material which may substitute cement in the production of concrete within the limits set by the cement norm EN 197-1 (see Table 1). According to that norm, cement may only be partially substituted by ground granulated blast furnace slag in the production of concrete. For CEM II, the mostly commonly sold cement type in Croatia, up to 35% of the final product may be other

(418) The case of imperfect transparency is consistent with the presence of positive margin opportunities for competitors in the relevant markets.

(419) Paul Klemperer, "Bidding Markets", UK Competition Commission discussion paper, June 2005, Section 4 (http://www.nuffield.ox.ac.uk/users/klemperer/BiddingMarkets.pdf).

(420) Horizontal Merger Guideline, paragraph 68.

(421) ID444, Form CO, paragraphs 220 and 438.

(422) Confidential ID719, ArcelorMittal's reply of 14 September 2016, which was made available to the Notifying Parties in the data room as set out in recital (8).

constituents than Portland cement. Buyers of cement have confirmed that they do not substitute grey cement with ground granulated blast furnace slag for failure to meet the technical requirements, for example on minimal cement content in concrete, or concerning setting speed.

(423) Second, while the Fortis Group is currently considering whether to build with CAMCE, a state-owned Chinese investor, a new cement plant in Novi Grad, Bosnia-Herzegovina no final decision to invest has been taken, the project timeframe is not determined, and the financial and business plans are not drawn-up yet, including for export of cement .

(424) Third, regarding the alleged intention of the Italian-based Fassa Group to start construction work for a new cement plant in Obravac (Dalmatia) with an annual capacity of 850kt:

(a) it is unclear if and when the required permissions for the construction of the plant will be granted;

(b) it will take around three additional years for the plant to be operational once permissions have been granted and construction has started; and

(c) the plant is intended to partly serve Fassa's own cement needs in Italy that are currently not served by Croatian suppliers.

(425) Fourth, Cementir Holding, a cement producer based in Taranto (Italy), does not sell cement in Croatia and is currently not interested in entering the relevant markets. Moreover, exporting into Croatia would only be considered by Cementir Holding only if it could profitably reach the local market from Italy and would provide an adequate return for the incremental investment required.

(426) Fifth, Sacci, a cement producer headquartered in Rome (Italy), does not sell cement in Croatia and is currently not interested in entering the relevant markets. Moreover, its cement assets are in the process of being acquired by Cementir .

(427) Sixth, Buzzi, a cement producer headquartered in Casale Monferrato (Italy), used to supply cement to customers in Croatia in limited volumes from its plant in Travesio, the closest one to Croatia. The plant was, however, definitively closed in June 2015 due to internal restructuring following the strong decline in Italian demand and Buzzi has had no sales in Croatia since July 2014. In addition, due to its current production and distribution network, Buzzi does not consider the relevant markets to represent a real business opportunity. Moreover, in order to supply significant quantities of grey cement in the relevant markets, Buzzi would have to set up a new production facility or distribution terminal.

(423) See customers' replies to the request for information of 16 November 2016.

(424) ID444, Form CO, paragraph 304.

(425) ID967, minutes of the telephone call with Kozaraputevi a.d. Fortis Group of 13 September 2016; ID2249, e-mail response from Grupa Fortis of 29.11.2016; ID2241, e-mail response of CAMCE of 14 September 2016 and ID2240 e-mail response of CAMCE of 29 November 2016.

(426) ID1984; additional submission 30th November by HeidelbergCement.

(428) Seventh, Colacem (plants in Italy and Albania), CRH (plant in Serbia), M.P.B. d.o.o. (Croatian trader) and Limak (plants in Turkey) currently have no grey cement activities in Croatia and are not interested in entering the relevant markets.

(429) Eighth, Albanian suppliers currently have no grey cement activities in Croatia and the Commission did not find any indications within its investigation that market entry by Albanian suppliers is likely. In addition, concerning market rumours picked up by Cemex Croatia that Albanian producers were seeking terminals for bulk cement imports into Croatia, Cemex Croatia commented in July 2016 that “[content of internal documents]”. As regards Albanian supplier Seamant in particular, Cemex underlines its shortcomings as a competitor as follows, calling into question their ability to constrain the Parties in the relevant markets:

(430) Ninth, as regards the activities of Colacem in Montenegro and potential sales from its Zelenika terminal, it is unlikely that such a terminal will be operational in the near future. This is for the following reasons:

(a) Colacem sold its shares in Eurocem Trade d.o.o., a joint-venture with Provadlex d.o.o which owns the terminal in Zelenika (Montenegro), in 2005. Therefore, "the Zelenika terminal project must be considered substantially aborted for Colacem S.p.A. […] The never defined legal disputes and controversies have been abandoned some years ago because no positive decision seemed to be obtainable.".

(b) Colacem further submits that "Therefore, the project has been abandoned by all parties for the future. The reason for the abandonment were legal disputes and controversies that did not promise a positive outcome that would have allowed the operation of the terminal as originally planned."

(c) According to publicly available information Provadlex' shares in Eurocem Trade d.o.o. are subject to legal restrictions.

(d) Provadlex' as well as Eurocem Trade D.O.O.'s capital have been subject to a freeze or are subject of similar financial limitation for more than 200 days and the Commission has been unable to contact Eurocem Trade d.o.o. or its shareholders. The Commission therefore doubts that those companies have currently any commercial activity.

(e) While the Zelenika terminal appears to have been put on sale in 2014 pursuant to an Order by a Montenegro Court, according to Titan's best knowledge "no potential buyer was identified".

(f) When asked whether in its view it is (i) possible and (ii) likely that the Zelenika Terminal in Montenegro could be used for the import of cement into Croatia in the near future, Titan explained: "As far as we are able to know, the Zelenika Terminal in Montenegro has never been put to use since its construction in 2008. There seems to be a lot of controversy between the owners (Eurocem Trade) and the local community regarding the legality of the construction permit issued for this terminal, based on the local community’s claims that the area where the terminal is built is not intended for industrial use pursuant to the planning regulations. We have no substantial knowledge of the legal circumstances surrounding this case, but we believe that unless such a dispute is finally and clearly settled, no investor is likely to assume the risk of putting this terminal into operation." As explained in 7.7.1

7.10. Countervailing buyer power

(431) Contrary to the Parties' submissions, the Commission concludes that even larger customers will have insufficient countervailing buyer power to prevent price increases by the merged entity after the Transaction for the following reasons.

(432) First, customer threats to sponsor entry in the relevant markets would be unlikely to eliminate the risk of price increases after the Transaction as attempts by larger customers to sponsor entry would likely imply higher costs for them as alternative suppliers would be located further away.

(433) Second, Cemex and DDC are not dependent on individual customers to an extent that would provide large customers with countervailing buyer power. The single largest customer of Cemex in Croatia accounts for […]% of Cemex's total sales whereas DDC's single largest customer in Croatia accounts for […]% of DDC's total sales (and […]% of DDC Kakanj's sales). The loss of even the largest Croatian customer would hence [strategic information].

(434) Third, any countervailing buyer power that larger customers may have held pre-merger due to the option to switch to from one of the Parties to the other will no longer exist because of the removal of the constraint exerted by DDC on Cemex and vice-versa.

7.11. Threat of actions by the merged entity further deterring entry or expansion by competing suppliers of grey cement

(435) For the reasons set out in recitals (436) to (442), the Commission concludes that the incentives of competing suppliers of grey cement to enter, or expand in the relevant markets is curbed by possible retaliatory actions by the merged entity concerning (i) the targeting of specific customer groups of the would-be entrant; and (ii) litigation strategies aimed in particular at importers. Such actions may result in making expansion of output by competing supplier of grey cement less profitable and more difficult.

(436) First, the merged entity would be able to detect quickly potential entry or expansion because the relevant markets are characterised by a degree of transparency both as regards pricing and the customer base of suppliers:

(a) according to one customer, "[i]t is a small market and everything is known";

(b) according to the Parties' internal documents, [content of internal documents];

(c) the Parties' sales staff perform [content of internal documents];

(d) according to customers that responded to the market investigation, information on competing price offers for grey cement is often disclosed in the context of commercial negotiations.

(437) Second, past behaviour suggests that both Cemex Croatia and DDC have resorted to actual or potential retaliatory actions to deter the threat of entry, either as a matter of general commercial strategy, or in specific reaction to certain conduct of competing suppliers of grey cement. Considering that such actions targeted customers in the would-be entrant’s sales area close to its production plant where the would-be entrant’s margins are typically highest, such actions can effectively curb its incentives to compete by expanding sales closer to the Parties’ plants.

(438) In the first place, DDC has sought to protect its market position in Bosnia-Herzegovina by actual or potential retaliatory actions against Cemex, Nexe and W&P:

(a) In reaction to the [internal document] by Cemex and an Albanian producer, the CEO of DDC Kakanj proposed in 2015 to […], board member of HeidelbergCement, to enter Montenegro [content of internal document]. In

(b) In 2014, DDC reacted to Nexe's increased imports into Hungary and Bosnia, as exemplified by [content of internal document]. Nexe itself confirmed that “DDC acts aggressively on the market” and observed that “in reaction to Našicecement’s [Nexe’s] exports, DDC targets [Nexe’s] customers in Slavonia and tries to win them over.” DDC's actions were detected and documented in a document by Cemex from 2014, which observed that [content of internal documents and strategic information].

(c) In 2014, DDC reacted to Cemex's ready-mix concrete pricing in Bosnia-Herzegovina by adapting its prices in Dalmatia: [content of internal documents].

(d) In 2015, DDC was preparing to enter Slovenia in reaction to W&P’s sales of bagged grey cement in Bosnia-Herzegovina from its plant in Anhovo, Slovenia: [content of internal document].

(439) In the second place, Cemex has sought to protect its market position in Croatia by actual or potential retaliatory actions against LafargeHolcim, Titan and DDC:

(a) In 2014, Cemex assessed the business case for potentially acquiring Asamer's grey cement plant in Bosnia, and considered that one of the main advantages of such acquisition was the [content of internal document];

(b) The same Cemex document analyses the competitive threat from Titan, observing that Titan was [content of internal document]. The analysis ends with the observation: “[content of internal document].”

(c) In 2009, Cemex retaliated against ready-mix concrete suppliers that were importing big bags (1.5t) of grey cement from Turkey into Dalmatia [content of internal document]. Moreover, DDC [content of internal document].

(440) Third, Cemex Croatia launched a series of legal proceedings against Italcementi in an attempt to curb sales of grey cement imported by Italcementi and its distributor A.R.M.A.C.O.M. The legal proceedings alleged a breach of Cemex’ Tecnocem brand registered in Croatia by Italcementi’s products in the period 2012-2013, which coincided with Croatia’s accession to the EU:

(a) Cemex’s claims against Italcementi gave rise to investigations by the Croatian Customs Administration, the Croatian State Inspectorate, the police department for commercial crime offences and the State Attorney’s office,

(b) In the follow-up legal actions for unfair competition filed by Italcementi and A.R.M.A.C.O.M., a judgment dated 21 December 2015 found that Cemex sought trade mark protection in bad faith: “the defendant is abusing its temporary monopoly as the brand holder and by registering the brand in bad faith, restricts competition between the parties by excluding the applicant from the Croatian market, or rendering access significantly more difficult by registering brands which the applicant has been itself using in Italy already for decades.”

(c) A judgment in an unrelated case involving Cemex referred to the judgment of 21 December 2015 and qualified Cemex’s actions as being “in breach of EU rules on competition and free movement of goods and capitals.” In this context, the court went on to observe that “[n]o company can be allowed to abuse in different ways and in such a grave form the entire legal system”

(d) Cemex Croatia's conduct had an impact on Italcementi and A.R.M.A.C.O.M. A.R.M.A.C.O.M. had difficulties to secure sales of Italcementi's grey cement in Croatia and reported a 30% decrease of its sales of bagged cement in Croatia during the first four months of 2013, as compared to sales in the same period of 2012. According to A.R.M.A.C.O.M., this decrease was partly due to Cemex Croatia's actions. A.R.M.A.C.O.M. was forced to seek an alternative source of grey cement supplies.

(441) Fourth, suppliers of grey cement in the relevant markets take into account the fact that sales in the traditional strongholds of competitors may result in actual or potential retaliatory actions:

(a) Correspondence between the […] and […], the board member responsible for the region, concerning DDC’s plans [content of internal documents and strategic information].

(b) According to one importer, potential suppliers of grey cement were generally unwilling to support imports into Croatia for fear of retaliation: “In search for alternative suppliers, MPB d.o.o. contacted a cement producer in Italy, which also had a cement production facility in Egypt and did not want to supply RMX plants in Dalmatia because it feared the reaction of Cemex. In responding to a request for a quote for cement supplies, Holcim Koromačno and Nexe would generally state that they had insufficient volumes. Tvornica cementa Kakanj [DDC] offered cement at a price that was not competitive. The only supplier ready to compete with Cemex in Croatia was DENİZLİ ÇİMENTO T.A.Ş from Turkey. Turkish cement is considered as the only option for imports as there is limited foreign ownership there and the local producers are not afraid of Cemex’ reactions in Turkey.”

(c) Asamer’s geographical footprint in Bosnia-Herzegovina and Croatia suggests a reluctance to enter the traditional markets of Cemex and DDC in southern Bosnia-Herzegovina. Asamer is located in the north-east of Bosnia-Herzegovina ([strategic information]) where it enjoys a regional market share of around [70-80]%. By contrast, in the neighbouring region [strategic information] to the south, Asamer has a market share of only [5-10]%, compared to [70-80]% of DDC. Further south, [strategic information], the Parties’ main overlap area in Bosnia, Asamer is absent (0% market share).

(442) Fifth, post-Transaction, the ability of the merged entity to undertake retaliatory actions would be even stronger:

(a) With the TCK plant in Kakanj, it will be easier for the merged entity to undertake retaliatory actions against Titan's sales in Dalmatia whereas Cemex previously considered that [content of internal documents and strategic information].

(b) Italcementi’s grinding station in Trieste, Italy, could be used for actions against importers into Croatia from Italy and Slovenia.

(443) Sixth, the conclusion that the incentives of competing suppliers of grey cement to enter, or expand in, the relevant markets would be curbed by possible retaliatory actions by the merged entity is not affected by HeidelbergCement's claim that the

Parties' market behaviour described in recitals (436) to (442) is "nothing else but vigorous competition."

(444) In the first place, the internal documents of the Parties acknowledge that the objective of retaliatory actions is not to ensure vigorous competition but rather to protect the prevailing price levels.

(445) In the second place, and in any event, the expectation of vigorous competition would also curb the incentives of competing suppliers to enter, or expand in, the relevant markets.

7.12. Impact of the Transaction

(446) For the reasons set out in recitals (447) to (474), the Commission concludes that the Transaction will lead to quantifiable price increases for grey cement and ready-mix concrete. This is based on the (i) contemporaneous internal documents of the Parties (section 7.12.1) as well as (ii) the Parties’ analysis of price effects prepared in the context of the Commission’s investigation (section 7.12.2). That evidence is corroborated by (iii) views expressed by a number of customers and competitors of the Parties (7.12.3).

7.12.1. Internal projections by DDC

(447) First, contemporaneous documents prepared by top management of DDC prepared in tempore non suspecto indicate that the Transaction would lead to price increases for grey cement and ready-mix concrete:

(a) A presentation entitled "Project Cerberus Group Review" dated 5 June 2015 analyses, amongst others, the impact of the Transaction on the competitive landscape in the Western Balkans region and states that the [content of internal document on pricing power after the Transaction].

(b) On 12 June 2015, [name] the CEO of Tvornica cementa Kakanj (DDC Bosnia), made the following observations: [content of internal document on pricing power after the Transaction].

(c) On 16 June 2015, [name] confirmed that the prices for cement were expected to increase: [content of internal document on price increases after the Transaction].

(d) Points 3 and 4 of a table dated created in May 2015 entitled “Synergy” (Figure 24) detail a [content of internal document on price increases after the Transaction].

Figure 24: Table "Synergy" dated May 2015

[…]

(475) ID2410, HeidelbergCement’s reply to the Statement of Objections, paragraphs 153-158.

(476) ID437-28, M.7878_Annex_RFI 20160711_QP7_Balkans Strategy 2014 draft v2.0 (CX 172).PDF, p. 38: "Increased retaliation opportunities [are] safeguarding prices."

(477) ID225-10386, M.7878 - QP 2 Section B (E-mails request) - USB Stick - GL_C001_00015250, p. 5 and 6.

(478) ID225-4053, M.7878 - QP 2 Section B (E-mails request) - USB Stick - GL_C001_00008256.

(479) ID225-3944, M.7878 - QP 2 Section B (E-mails request) - USB Stick - GL_C001_00008052.

(480) Based on document metadata.

(481) ID1103-975, Response to questionnaire 2 of 28 October 2016. - M7878_All_files - GL_C001_00027199.

(482) ID225-9623, M.7878 - QP 2 Section B (E-mails request) - USB Stick - GL_C001_00022091, ID1088-975.

(483) ID225-11579, GL_C001_00027199.pdf.

(484) [Publically unavailable and internal information].

(485) Figure 25: Slide "Croatia Strategic fit", "Project Cerberus", dated 23/24 June 2016

[…]

(486) The implied average gross revenue per ton in 2016 is […]€/t for DDC and […]€/t for HeidelbergCement.

(487) Second, the Parties have failed to put forward any convincing evidence to explain the multiple indications of the price increases in the contemporaneous documents prepared by top management of DDC, as referred to in recitals (447) to (449).

(488) In the first place, the Parties offer unconvincing ex post facto alternative interpretations for the various quotes reproduced in recital (447) to claim that the references to “increased pricing power” or “price increases” should be understood as not having their usual meaning.

(489) In the second place, the Parties' argument that the EUR […] million of "logistics optimisations" as reproduced in Figure 25 are not a disguised price increase but relate to a reduction in distribution costs for deliveries to customers in Bosnia-Herzegovina and Dalmatia is unconvincing:

(a) the sums allegedly representing “logistics optimisation” (EUR […] million) are by and large consistent with the additional revenues from the price increase set out in earlier internal documents (EUR […] million );

(b) contrary to the Parties' argument that increased revenues may simply be the result of incomplete pass on of logistics savings to customers, the slide in Figure 25 indicates an expected increase in delivered prices to customers: the sum of EUR […] million is categorised as "gross sales" and the implied average pre-merger prices per ton in the document […]€/t for DDC and […]€/t for HeidelbergCement in 2016) are in line with delivered prices to customers rather than ex-works prices.

(490) In the third place, the magnitude of EUR […] million of logistics costs savings, that is to say a reduction in transport costs as the merged entity would serve customers from the closest of its plants, is implausible:

(a) The vast majority of the Parties' sales are [strategic information]. Therefore, transport cost savings could […] apply to […] of the Parties' volumes. For example, Cemex' customers in Croatia and in Bosnia-Herzegovina who are geographically closer to DDC's Kakanj plant than to Cemex's Split plant represent about […]kt of Cemex sales in 2015. In contrast, the figure of EUR […] million is the result of applying savings of […]€/t to a volume of […]kt which is allegedly made up of DDC's total sales in Croatia (~[…]kt) on the one hand and Cemex' total sales in Croatia (~[…]kt) and Bosnia-Herzegovina (~[…]kt) on the other hand.

(b) The synergy table of May 2015 quoted in Figure 24 suggests that expected logistic savings concern more limited sales volumes. The figure indicates that

(487) ID2410, HeidelbergCement reply to the Statement of Objections, Section 4.1.

(488) ID2410, HeidelbergCement reply to the Statement of Objections, paragraph 168-170.

(489) [Publically unavailable and internal information].

(490) ID2410, HeidelbergCement reply to the Statement of Objections, para 173.

(491) See Figure 24.

(492) ID2410, HeidelbergCement reply to the Statement of Objections, para 173 ID2901, HeidelbergCement reply to the Letter of Facts, para 53.

(493) Computation based on Cemex's data. See Response to RFI of 28 October 2016, and subsequent data RFIs.

(494) ID2596, HeidelbergCement Response to RFI, 19 January 2017, page 2.

(495) In addition, a […]€/t transport cost saving would correspond to a reduction of average transport distance by […]km on all affected volumes if the Parties' estimate of transport costs of […] EUR/km is applied. The computation follows the same reasoning applied by the Parties in their maximum effect analysis (ID437-255 Annex 6.III.1.-Grey Cement_HR_E.CA economic analysis_revised, p. 30-33).

(95)

(c) An email from [name] of 8 July 2015 (10.16am) equally refers to lower volumes which would be subject to logistics costs savings and hence implies much lower logistics costs savings: [content of internal document].[Name] hence expected logistics savings on […]kt amounting to only EUR […] million in total.

(455) In the fourth place, in response to a request for information of the Commission, the Parties did not provide a convincing explanation as to why logistics costs savings would be achieved on all sales volumes of the Parties. The Parties only noted that the estimate was based on a "wider interpretation" of the term "logistic optimisation" which would cover "transport costs, cost reduction initiatives, working capital product portfolio alignment, optimization of production, etc." in addition to "pure logistic savings". This argument is unconvincing as it is in contradiction with the Parties' explanations elsewhere that logistics optimisations would come from reduced transport costs by supplying customers from the closest plant. Furthermore, the Parties' submissions on those alleged other cost savings remained unsubstantiated.

(456) In the fifth place, the internal HeidelbergCement email of 8 July 2015 does not lead to a different conclusion:

(a) the email was prepared by members of HeidelbergCement’s M&A department that had limited knowledge of the relevant markets; in contrast, the staff originally projecting the price increases included the CEO of DDC TCK who personally and on a daily basis oversaw the pricing policy of DDC;

(b) price increases were still implicit in the financial data in the aforementioned slide entitled “Croatia Strategic Fit” (see Figure 25) as that slide projects increases in gross sales without increases in sales volumes, which can be explained only by price increases, as explained in recitals (448)(b) and (449). This slide was still used a year after the email of 8 July 2015 in a presentation dated 23 June 2016. The explanations of HeidelbergCement's counsel during the Oral Hearing that the document of 23 June 2016 is irrelevant as it was a mere copy of the slide used in the presentation of 23/24 June 2015 are unconvincing. The slides used in 2015 and 2016 were not identical as the 2016 version contained a slight downward modification of the expected additional revenues for cement ([…]m EUR to […]m EUR) and RMX ([…]m EUR to […]m EUR).

7.12.2. The Parties' analysis of price effects

(457) The Parties submitted an analysis according to which the maximum price effect of the Transaction for customers in Dalmatia would on average be [0-5]% for bulk grey cement (within a range of [0-5]%-[0-5]%) and [0-5]% for bagged cement (within a range of [0-5]%-[0-5]%). The analysis is based on the "next-closest" competitor to each customer in Dalmatia. The change in the competitive constraint on the merged entity was approximated by the additional transport costs that the next closest competitor would incur to serve the customers. The Parties consider this additional transport costs to approximate the maximum price effect that could arise to the customer.

(458) For the reasons set out in recitals (459) to (468), the Commission concludes that the Parties' analysis does not provide a reliable measure of the likely maximum price effect of the Transaction.

(459) First, the Parties' estimate of maximum price effects is directly contradicted by their internal projections (see recitals (447)-(450)).

(460) Second, the Parties' analysis focuses only on transport costs, thereby underestimating the competitive disadvantage of competitors arising from additional distance to the customers. While transport costs are an important parameter of competition and an important driver of costs in the relevant markets, other important factors such as security of supply, quality and reputation (see Section 7.7.1) are not captured by the Parties' analysis. For example, security of supply considerations are likely to increase with the distance to a customer, such that if the distance increases, security of supply issues become more severe. Therefore, the Parties' argument that security of supply is implicitly included in the analysis because such considerations apply also to DDC cannot be accepted.

(461) Third, the Parties' analysis contains a number of methodological errors that lead to an underestimation of the transport cost disadvantage of the next best alternative supplier.

(462) In the first place, the Parties' analysis ignores that the next closest competitor may not be able to serve a given customer due to capacity constraints. This is the case for example of LafargeHolcim's terminal in Zadar that is identified as the geographically next-closest alternative supplier after DDC for a significant number of customers, yet is currently facing capacity constraints, as set out in recitals (328) to (332).

(463) In the second place, the Parties' analysis initially included plants that either have never been opened or are now closed or mothballed. Most notably, the Parties included Colacem's terminal in Zelenika, Montenegro, among possible suppliers. That terminal has however never been completed and is unlikely to be opened in the foreseeable future, and should therefore be excluded from the analysis.

(464) In the third place, the Parties' analysis includes oversea suppliers that will be at a particular competitive disadvantage compared to the merged entity (see section 7.7.4).

(465) In the fourth place, when calculating transport costs via terminals, the Parties' analysis only accounts for the transport costs from the terminal to the customer and not transport costs from the plant to the terminal. The Parties' argument that transport costs from plant to terminal should not be included because terminals are continuously resupplied cannot be accepted. Transport costs from plant to terminal are part of the incremental costs of selling additional volumes through the terminal in the same way as variable production costs or transport costs to reach the final customers.

(466) Fourth, for a large majority of customers in Dalmatia, the geographically next closest alternative supplier after DDC indicated by the Parties analysis is a competitor that is currently not active in that region. For example, for 55-60% of customers in the Parties' analysis as modified by the Commission the geographically next-closest supplier is Asamer, and for other customers the geographically next-closest competitor is a supplier from Albania.

(467) Fifth, the Parties' analysis is based on the assumptions that customers would be able to play the two geographically closest competitors perfectly off against each other and that distance is the only relevant differentiator between suppliers of grey cement so that the customer sources cement from the geographically closest supplier at a price equal to the marginal cost of the next closest supplier. This is, however, not observed in practice. For example, DDC's sales to Dalmatia are made at a significant margin despite customers in Dalmatia being closer to Cemex' Split than to Kakanj. This indicates that competition for individual customers is not as intense as the Parties' model suggests.

(468) Sixth, correcting the Parties' analysis leads to significantly higher estimated transport costs disadvantages for the geographically next closest competitors to customers in the relevant markets after DDC. The Commission adjusted the Parties' analysis by excluding overseas suppliers, by excluding closed plants and by correcting the transport costs for the terminals for the reasons discussed in recitals (461) to (465). The revised Parties' analysis leads to estimated transport cost disadvantages of the geographically next-closest supplier of, on average, [0-5]% (with a maximum of [5-10]%) of DDC's average price in Dalmatia for bulk cement, and [0-5]% (with a maximum of [5-10]%) for bagged cement. This illustrates that even in the context

(505) ID2410, HeidelbergCement reply to the Statement of Objections, Annex 1, paragraph 170c.

(506) Such suppliers are identified in the Parties' analysis as the next best alternative supplier for […]% of the bagged cement customers.

(507) ID2410, HeidelbergCement reply to the Statement of Objections, Annex 1, paragraph 170d.

(508) The Parties' analysis essentially assumes an open outcry auction format where the geographically closest supplier wins a customer at the marginal costs of the geographically second closest supplier and where transport costs represent the only difference in competitive position of different suppliers.

(509) In its reply to the Statement of Objections, Annex 1, paragraph 172, HeidelbergCement argued that the results of the model as modified by the Commission differ from the ones presented by the Commission in the Statement of Objections, and are actually lower ([0-5]% on average for bulk cement and [0-5]% on average for bagged cement). The Commission notes that the discrepancy is due to a different weighting applied. Specifically, the figures reported by HeidelbergCement in its reply to the Statement

(98)

(469) The Commission considers that the evidence from the Parties’ internal documents (section 7.12.1) and the price effects analysis (7.12.2) is further corroborated by the views of a number of customers and competitors of the Parties.

(470) First, certain customers submit that the Transaction would further strengthen the market power of Cemex Croatia and likely result in price increases due to decreased competition, with LafargeHolcim remaining as the only sizeable competitor. They also claim that for certain parts of Dalmatia (the stretch from Split to Dubrovnik), DDC would be removed as the only alternative to Cemex:

"The effect of the transaction for Strabag in the region of Dalmatia: High risk of increasing the price of cement and more difficult competition in the market of ready-mix concrete with no alternative supplier of grey cement. In our view there will be a less competitive environment on the market for grey cement in the region of Dalmatia in the end. In our view there will be only a minor effect of the transaction on our company in the view of the supplier of grey cement." "We expect a limitation of supply of cement on two supplier (rather three till now) which is negative in terms of competition." "The effect of the transaction on the markets of ready-mix concrete in the Croatian region [Dalmatia] will be negative, as there will be no alternative supplier of grey cement."

"Considers that following the merger the prices will be increasing and not falling. It will have the supply of cement, but the price at which the cement will be sold to it is questionable. Declares that in the event that Cemex and Kakanj increase prices, it has to continue buying from them and has no other alternative. If cement prices increase consequently the price of concrete it produces will increase as well."

"States that Cemex holds high prices and that after the merger of Cemex and Kakanj, they will, as supplier, hold the area of Hercegovina and South Dalmatia. To the best of its knowledge CMA is the southernmost [Imotski] customer of Holcim cement. In the area towards Dubrovnik, the deliveries are made only by Cemex and Kakanj."

"I consider we will not [have sufficient alternative suppliers of cement] considering that on our market Cemex and DDC hold the whole supply of cement", " We are not sure [whether we will have the same or similar conditions for supply of grey cement for our production of RMX concrete], we believe the prices will increase." [Effects of the concentration] It all depends on the policy they will take, at this moment it is very difficult to assess the effect. By implementation of proposed transaction it is

of Objections were simple averages, whereas the figures reported by the Commission are volume weighted averages. In the Commission view, volume weighted figures are more relevant as they assign a higher weight to larger customers and a lower weight to smaller customers. The maximum effect for bagged cement reported in the Statement of Objections was, however, inaccurate ([5-10]%); the Commission corrected the figure to [5-10]%.

(99)

(470) The Commission considers that the evidence from the Parties’ internal documents (section 7.12.1) and the price effects analysis (7.12.2) is further corroborated by the views of a number of customers and competitors of the Parties.

(471) Second, some customers consider that given the oligopolistic nature of the market, the Transaction may also result in price increases from reduced competitive pressure on third party suppliers, such as LafargeHolcim.

"Our current supplier HolcimLafarge is already an alternative supplier therefore we consider that we can have an alternative supply of cement, but we are sceptical regarding the price of cement considering there will be one competitor less on the market." and "[Regarding the conditions for supplying of grey cement for the production of concrete] We expect the increase of the price of cement."

"Merger means reducing the total number of enterprises while reducing competition and the effect is often negative and reflected in an increase in prices of cement and thereby weakening the position of our company in relation before merger. The negative effects of that often can be paying higher prices for cement with reduced quality. As a construction company and as an important purchaser of cement, we need a quality product, safe and timely delivery and competitive prices."

(472) Third, domestic competitor LafargeHolcim considers that the Transaction will strengthen Cemex and DDC's positions in the Croatian cement markets. It is claimed that the Transaction "will likely make it more difficult to compete in Dalmatia (and especially Bosnia and Herzegovina) due to [our] high transportation costs as outlined above – the transaction will lead to increased supply from Split and Kakanj together which will lead to the combined entity having a very significant share in Dalmatia which will be difficult for [us] to compete with" and "will have a strong influence in all regions in Croatia – this is particularly so when you consider the levels of cement production that Heidelberg has in neighboring countries of Italy, Hungary and Bosnia, from which it can and does import into Croatia. We expect therefore that Heidelberg will become an increasingly significant producer, especially when its production capacity across these countries and within Croatia is

(473) On the other hand, a number of operators contend that the Transaction might result in lower prices from the merged entity due to logistic savings, broader offer etc. Those expectations however are frustrated by the Parties’ declared intentions to increase the prices and not to pass-on any efficiencies to customers (see section 7.12.1) and the Parties' maximum effects analysis (see section 7.12.2).

(474) In addition, a DDC document indicates that, during the review of this transaction by the Bosnian Competition Council, customers have sought and received input from DDC on how to answer to queries by the Bosnian competition authority. An email from the CEO of TCK to Schwenk explains that [internal document].

8. SIGNIFICANT IMPEDIMENT OF EFFECTIVE COMPETITION IN A SUBSTANTIAL PART OF

THE INTERNAL MARKET

(475) Pursuant to Article 2(2) and (3) of the Merger Regulation, a concentration "which would not significantly impede effective competition in the common market or a substantial part of it […] shall be declared compatible with the common market." and a concentration "which would significantly impede effective competition in the common market or a substantial part of it […] shall be declared incompatible with the common market."

(476) One of the pre-conditions for a finding of incompatibility is therefore that the concentration will significantly impede effective competition or in a substantial part of the internal market. It follows that where the Commission identifies a significant impediment to effective competition, it can only make its approval conditional upon remedies aiming at solving such concerns if the significant impediment relates to a market constituting a substantial part of the internal market.

(477) In reply to the 6(1)(c) Decision, the Parties' submitted arguments that the region of Dalmatia or a 150 km circular catchment area around Split cannot be considered to be a substantial part of the internal market as the areas concerned are very limited and "at the very edge of the EU".

(478) As explained in section 6.3, in its in-depth investigation the Commission has further analysed the geographic scope of the relevant markets. For the reasons set out in that section, the Commission has both extended and refined the relevant catchment area as a proxy to define the relevant geographic market. The market definitions retained are the 250km circular catchment area around the Split plant and the modified 250km catchment area around the plant. The circular 250km circular catchment area approach corresponds to the market definition suggested by the Parties (see section 6.3.2).

(479) Each of the alternative relevant markets is a sizeable one, with the surface of the underlying region exceeding 30,000km and accounting for more than 2 million inhabitants. Its current yearly cement consumption of […] and […]kt respectively

(480) The Commission considers that each of the alternative relevant markets constitutes a substantial part of the internal market not least because of its size in terms of cement demand, the population living in it and the exports from those markets into other EEA countries.

9. COMMITMENTS

9.1. Principles

(481) Where the undertakings concerned modify a notified concentration, in particular by offering commitments with a view to rendering the concentration compatible with the internal market, the Commission should be able to declare the concentration, as modified, compatible with the internal market. Such commitments should be proportionate to the competition problem and eliminate it entirely.

(482) Under the Merger Regulation the Commission has the power to only accept such commitments that are capable of rendering a notified concentration compatible with the internal market.

(483) As set out in the Remedies Notice, the commitments have to eliminate the competition concerns entirely, and have to be comprehensive and effective from all points of view.

(484) In assessing whether commitments will maintain effective competition, the Commission considers all relevant factors, including the type, scale and scope of the proposed commitments, with reference to the structure and particular characteristics of the market in which the Transaction is likely to significantly impede effective competition, including the position of the Parties and other participants on the market.

(485) The commercial structures resulting from the commitments must be sufficiently workable and lasting to ensure that the significant impediment to effective competition will not materialise. Moreover, commitments must be capable of being implemented effectively within a short period of time. The Commission may not be able to conclude that it will be possible to implement the commitments if there are third party rights or if there is a risk of not finding a suitable purchaser.

(486) As the basic aim of commitments is to ensure competitive market structures, structural commitments are as a rule preferable over behavioural commitments. Furthermore, divestiture remedies are the best way to eliminate competition concerns resulting from horizontal overlaps. Other structural commitments – such as granting

Calculated based on the Parties' estimates submitted with ID1280 Answer Q58 of M.7878_Response to Art 11 RFI_20161028_consolidated response_9 Nov 2016.PDF.

Merger Regulation, Recital 30.

Judgment of 25 March 1999, Gencor v Commission, T-102/96, EU:T:1999:65, paragraph 318.

Commission Notice on remedies acceptable under Council Regulation (EC) No 139/2004 and under Commission Regulation (EC) No 802/2004 (OJ C 267, 22.10.2008, p. 1-27).

Remedies Notice, paragraph 12.

Remedies Notice, paragraphs 9-11.

access to key infrastructure – may be suitable to resolve concerns from horizontal overlaps if they are equivalent to divestitures in their effects.

(487) The most effective way to maintain effective competition is via divestiture 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 has to be viable as such. Therefore, the resources of a possible or even presumed future purchaser are not taken into account at the stage of assessing the remedy unless a sale and purchase agreement is concluded during the procedure or the commitments contain an upfront buyer clause as well as specific requirements as to the suitability of the buyer.

(488) Commitments proposed to the Commission pursuant to Article 8(2) must be submitted to the Commission within not more than 65 working days from the day on which proceedings were initiated. Where the deadlines for the final decision have been extended according to Article 10(3) of the Merger Regulation, also the deadline for remedies is automatically extended by the same number of days.

(489) As regards commitments that are submitted out of time, the parties to a notified concentration may have such commitments taken into account subject to two cumulative conditions, namely, first, that those commitments clearly and without the need for further investigation resolve the competition concerns previously identified and, secondly, that there is sufficient time to consult the Member States on those commitments.

9.2. The Commitments submitted by the Notifying Parties and the negotiations with Titan and Asamer

(490) On 26 January 2017, the Notifying Parties formally submitted commitments (the "Commitments").

(491) The Notifying Parties have committed to terminate an existing lease agreement between Cemex Croatia and the port of Ploče for a cement handling terminal in Metković in Dalmatia that still runs until the end of [year]. The terminal is located in the port of Metković which is part of the port of Ploče. The terminal is located about 25km inland on the Neretva river and does not have access to a deep sea port.

(492) The terminal is owned by the port of Ploče and is a storage facility for bulk and bagged cement with existing truck and vessel access. The terminal currently does not have a functioning railway access and there is no functioning equipment to load or unload trains carrying cement.

(493) Cemex Croatia uses the terminal for [amount], mainly to [countries]. In 2014, 2015, and 2016, Cemex sold […]kt, […]kt and […]kt of cement through the Metković terminal respectively achieving a turnover of EUR […] , EUR […]and EUR […]respectively, resulting in total losses of EUR […], EUR […]and EUR […]respectively.

(494) Pursuant to the Commitments, the terminal would be leased by a competitor ("new lessee") and the Notifying Parties would provide the new lessee with certain customer details and support regarding logistics to facilitate sales in Croatia. More specifically, the Notifying Parties have committed:

(a) to terminate the lease by Cemex Croatia and waive Cemex Croatia's [content of internal documents] (clause 2.1 of the Commitments);

(b) to find a suitable new lessee with the ability and incentive to effectively compete on a long-term basis with DDC in Southern Croatia and in particular

(c) to procure that DDC will:

(a) provide the new lessee with all customer records for Croatian customers directly supplied from DDC Kakanj in the last two calendar years (clause 2.2 (iii)(a) of the Commitments);

(b) provide logistic support to the new lessee by providing the contact details of transport companies used by Cemex Croatia and, at the option of the new lessee, by selling one of DDC's cement silo trucks to the new lessee (clause 2.2 (iii)(b) of the Commitments);

(c) maintain, at the option of the new lessee, a back-up facility of […]kt cement per year for the benefit of the new lessee at the Split plant which the lessee can use (i) at any time after giving at least […] days prior notice to DDC and (ii) at a price which is set by DDC each year, [pricing information], and which has been approved by the Trustee (clause 2.4 of the Commitments);

(d) not to implement the Transaction before the new lessee and the port of Ploče have entered into a final binding new lease agreement and the Commission has approved the new lessee and the terms of the new lease agreement (clause 2.3 of the Commitments);

(e) to give up the possession of the terminal, remove all cement and other materials and clean the terminal within […] days after closing the Transaction (clause 2.5 of the Commitments); and

(f) not to lease, use or acquire the terminal for […] years (clause 3 of the Commitments).

(495) At the time of the submission of the Commitments, Titan, with plants in Serbia and Albania, had entered into a non-binding Memorandum of Understanding concerning the lease of the terminal with DDC, Cemex Croatia, and the Port of Ploče.

(496) On 9 February 2017, Titan informed the Commission, however, that after further analysis of the details that Titan had received in the context of the negotiations with DDC, Cemex Croatia and the Port of Ploče, it had come to the conclusion that the lease of the terminal was not an attractive business opportunity.

(497) On 17 February 2017, the Commission informed the Notifying Parties that Titan had come to the conclusion that the lease of the terminal was not an attractive business opportunity. The Commission also informed the Notifying Parties that during the Commission's market test, Asamer expressed its interest in negotiating a lease of the Metković terminal.

(498) On 14 March 2017, the Notifying Parties informed the Commission that on 13 March 2017, Asamer had entered into an agreement with the Port of Ploče for the lease of the Metković terminal.

(499) On the same day, Asamer submitted to the Commission a copy of the signed agreement it entered into on 13 March 2017 for the lease of the Metković terminal.

Pursuant to Article 16(i) of the agreement, the lease is conditional on the Commission declaring the Transaction compatible with the internal market.

9.3. The Parties' submissions

(500) The Notifying Parties submit that the Commitments will eliminate the significant impediment to effective competition in the relevant markets by allowing Titan or any other suitable new lessee approved by the Commission to enter or expand its competitive position in Croatia (and in particular Dalmatia) via the Metković terminal within a short period of time.

(501) In the Form RM, the assessment of new lessees focused on Titan, with plants in Serbia and Albania, referring to the non-binding Memorandum of Understanding that had been signed between DDC, Cemex Croatia, Titan and the Port of Ploče. The Notifying Parties provided no specific information about Asamer in the Form RM.

(502) In the Form RM, the Notifying Parties also focussed on the supply of grey cement to the Metković terminal by road or sea, given that, according to the Notifying Parties, no investment in infrastructure for unloading/loading from/into trucks and sea vessels will be required.

(503) By contrast, in the Form RM, the Notifying Parties provided limited information on the supply of grey cement to the Metković terminal by rail :

(a) The Form RM described only the transport of grey cement by rail from the Metković terminal into Bosnia-Herzegovina and not how the Metković terminal could be supplied by rail from a potential lessee's production site.

(b) While the Form RM noted that the shipment of grey cement from the Metković terminal to customers by rail would require investments since the unloading facilities are inoperable, the Notifying Parties did not provide information in the Form RM on the investments in the railway infrastructure needed to ship grey cement from the new lessee's site to the Metković terminal.

(c) The Form RM provided only costs for rail supply from Titan's plant in Kosjerić/Serbia to the Metković terminal and no information on costs of rail delivery to the Metković terminal from the sites of other new lessees' locations.

(d) [Content of internal documents] by the existing lease contract between Cemex and the Port of Ploče and the Form RM provided no information regarding

whether and under what conditions the Port of Ploče would agree to such supplies.

(e) The Form RM also did not provide any justification for the Notifying Parties' [handling costs] estimate of the variable costs for handling cement shipped to the Metković terminal by rail.

(f) While the Form RM indicated that the effective annual capacity of the terminal for rail was […]kt, it did not explain how that capacity had been calculated and what the bottlenecks for the capacity utilisation were.

(g) The Form RM left open whether the capacity only referred to shipping cement to customers from the terminal or whether that same capacity could be achieved when shipping cement from the lessee's production site to the Metković terminal.

(504) On 23 February 2017, and after having been informed by the Commission that Titan had concluded that the lease of the terminal was not an attractive business opportunity, the Notifying Parties argued for the first time that Asamer was best placed to enter into a lease for the Metković terminal.

9.4. Responses to the market test of the Commitments

(505) On 1 February 2017, the Commission launched a market test of the Commitments.

(506) On 17 February 2017, the Commission informed the Notifying Parties about the responses to the market test of the Commitments, which concerned different aspects including the Metković terminal's operating costs, the importance of the origin of the grey cement sold through the Metković terminal, the capacity and security of supply of the Metković terminal, the role of customer relationships, the ease of expansion and entry through the Metković terminal and the identity and suitability of new potential lessees.

(507) In essence, customers and competitors that responded provided mixed feedback on the ability of a new lessee to compete effectively in the relevant markets. While respondents generally considered it possible to start selling grey cement through the Metković terminal through supplies from a plant other than Cemex' Split plant, the majority of respondents identified challenges for a new lessee to establish itself as a self-sufficient, effective and viable business in Croatia. These challenges included supply logistics, cost disadvantages vis-à-vis DDC and the absence of existing business relations. The Commission also informed the Notifying Parties that Asamer had expressed an interest in the lease of the Metković terminal.

(508) On 23 February 2017, the Commission discussed with the Notifying Parties the Commission's views on the Commitments and the results of the market test at a formal state-of-play-meeting.

(550) Those costs were estimated at […] EUR/t, ID2725, Form RM, table 4.

(551) Such information was later provided in response to the Commission's request for information of 16 February 2017, question 3 (ID3509). The Form RM had simply noted for the unloading capacity "[content of internal documents]" and for the rail loading capacity "[content of internal documents]", ID2725, Form RM, page 35.

(552) See presentation during the state-of-play meeting of 23 February 2017, ID3600.

(553) As set out in recital (17), the Parties were given access to the responses of customers and competitors to the market test of the Commitments on 20 February 2017.

9.5. The Commission's assessment

(509) For the reasons outlined in recitals (511) to (606), the Commission concludes that the Commitments do not eliminate the competition concerns entirely and are therefore insufficient to render the concentration compatible with the internal market.

(510) First, the Commitments suffer from a number of structural deficiencies (section 9.5.1). Second, there is a low likelihood of finding a suitable lessee (section 9.5.2). Third, the remedy would be insufficient in scale (section 9.5.3). Fourth, there are shortcomings in the modalities of the implementation of the Commitments (9.5.4).

9.5.1. Structural deficiencies of the Commitments

(511) The Commission concludes that the Commitments suffer from a number of structural deficiencies.

(512) First, the Commitments will have a limited impact on the market power of the merged entity.

(513) In the first place, while the merged entity would have sales market shares between [40-50]% and [50-60]% in the relevant markets with joint market shares in the region of Dalmatia reaching [70-80]-[80-90]% (see section 7.5.3), a new lessee would have no sales since it would not take over any existing customer relationships or contracts. Moreover, it is uncertain what sales levels a new lessee would achieve in the next two to five years. This was acknowledged by DDC's submission that it is "not in a position to predict Titan's or any other new lessees's future business strategy".

(514) In the second place, the Commitments would have limited effect on the capacity concentration levels in the relevant markets, and this even if the cement handling capacity of the Metković terminal were assigned to a new lessee in the same way as production capacity and even if the Notifying Parties' estimate for the capacity of the terminal of […]kt is used (for a discussion of the actual capacity levels see section 9.5.3).

(515) Figure 27 and Figure 28 illustrate the customer-based joint capacity shares of the merged entity and the capacity share increments after implementation of the Commitments (calculated in the same way as outlined in recital (292).

Figure 27: Joint capacity shares of the merged entity after implementation of the Commitments

[…]

Figure 28: Capacity share increments after implementation of the Commitments

[…]

(516) Second, it is uncertain whether the Commitments would have any lasting effect because the Notifying Parties have committed to an initial term of the lease for a […]-year period. Therefore, there is a risk that a new lessee might cease its operations at the latest in [year].

(517) Third, the Commitments would not lead to the divestiture of a self-sufficient, effective and viable business.

(518) In the first place, the Metković terminal is a storage facility not a production plant and can therefore not generate any grey cement sales on a stand-alone basis.

(519) In the second place, the Commitments also do not include any access to grey cement supplies from other sources.

(520) In the third place, the Commitments do not include any brands, customer relationships or other goodwill.

(521) In the fourth place, the Commitments do not include the transfer of any staff, such as managerial, sales or administrative staff.

(522) Fourth, the Commitments offer a mere, uncertain, business opportunity for a new lessee to start selling, or to expand sales of, grey cement in the relevant markets which is not comparable to the divestiture of an existing standalone business.

(523) In the first place, a new lessee would operate on the basis of a newly established business model, transporting grey cement from a source not previously used by the Metković terminal on the basis of newly established logistics routes.

(524) In the second place, the supply to Croatian customers from the Metković terminal would have to be newly established as [amount] sales have been made from the Metković terminal in the past three years.

(525) In the third place, the uncertainty of a new lessee growing into a viable competitive force that could compete effectively with the merged entity on a lasting basis was acknowledged by customers and competitors in their responses to the market test:

(a) The majority of customers expressing a view regarding the question "how easy or difficult it would be for a new lessee to establish a viable, competitive and sufficiently sizable cement business in Croatia and in particular in Dalmatia comparable to that of the DDC Kakanj business in Croatia", indicated that it would be hard or very hard.

(b) The majority of customers expressing a view regarding the question concerning the particular advantages and disadvantages the new lessee would face when competing for customers with Cemex Croatia and DDC indicated, that a new lessee would face disadvantages, for instance because DDC and Cemex were "the well-established players in the market" and they had "already built business relationships".

(c) The majority of customers expressing a view regarding the question whether "the provision of DDC's customer records is sufficient for the new lessee to establish customer relationships with customers of bulk and bagged cement in the near future (next three years)" indicated that a new lessee would need more customers and additional time to establish customer relationships.

(526) Fifth, while the new lessee would have to enter into a contract with the Port of Ploče on substantially the same terms as the existing contract between Cemex and the Port of Ploče, this would create uncertainty for any new lessee due to the lack of clarity regarding the interpretation of certain clauses of the existing contract, including the clauses concerning the variable costs of handling volumes over […]kt of grey cement. While the Port of Ploče considers that above […]kt, such variable costs are […] EUR per ton, Cemex considers, that such variable costs should, like below […]kt of grey cement, remain at […] EUR per ton even for volumes of grey cement above […]kt.

9.5.2. Low likelihood of finding a suitable lessee

(527) The Commission concludes that the likelihood of the Notifying Parties finding a suitable lessee for the Metković terminal is low.

(528) First, each new lessee of the Metković terminal identified by the Notifying Parties (Titan, Asamer, W&P, Cimko) would be less cost-competitive than DDC, regardless of the mode of transport (sea, truck, rail) used and those costs disadvantages would not be offset by security of supply considerations (section 9.5.2.1).

(529) Second, it is revealing that Titan has decided not to pursue the negotiation of the lease since Titan did not consider the lease of the Metkovic terminal to be an attractive business opportunity (section 9.5.2.2).

(530) Third, Asamer would not be a suitable lessee of the Metković terminal, in particular since it would have neither the ability nor the incentive to develop into a viable competitive force that could compete effectively with the merged entity on a lasting basis (section 9.5.2.3).

9.5.2.1. Cost disadvantages of potential lessees and security of supply considerations

(531) In order to develop into a viable competitive force that could compete effectively with the merged entity on a lasting basis, a new lessee would have to be similarly cost-competitive as DDC in serving customers in southern Croatia.

Market Test to Customers, replies to question 18.

Market Test to Competitors, replies to questions 27 and 29.

ID3140, non-confidential version of Port of Ploce's Response to RFI of 02 Feb 2017, question 11a.

Remedies Notice, paragraph 47.

The same conclusions set out in recitals (539) to (559) would be valid for other new lessees using seaborne, road-based or rail-based supplies as they are located at a similar or higher distance than the suppliers whose costs have been analysed. In any event, none of the other cement producers contacted has shown an interest in the lease of the terminal.

In the assessment of the Commitments in this section 0, southern Croatia is defined as the area of activity of DDC Kakanj in Croatia. This comprises Dalmatia and the southern part of Kvarner.

(532) The Commission has focused on the grey cement suppliers that showed an interest in leasing the Metković terminal in response to the Commission's market test: Titan from its plants in Kosjerić (Serbia) and in Fushe Kruje (Albania), Cimko from its plant in the Yumurtalık region in Anatolia (Turkey), W&P from its plant in Anhovo (Slovenia) and Asamer from its plant in Lukavac (Bosnia-Herzegovina).

(533) In reaching the conclusion that each new lessee of the Metković terminal that showed an interest in leasing the Metković terminal in response to the Commission's market test would be less cost-competitive than DDC, the Commission has compared the variable cost-to-market in southern Croatia of each new lessee to that of DDC, analysing all means of transport to supply the Metković terminal, namely, seaborne, road-based and rail-based transport (although the rail connection is not operable, as set out in recital (492), and was not presented as a means of transporting grey cement to the Metković terminal by the Notifying Parties in the Form RM as set out in recital (502)).

(534) Regarding the variable cost-to-market of DDC in southern Croatia, the Commission has computed and added together DDC's average transport cost to reach customers of […] EUR/t, and its average production cost of bulk cement of […] EUR/t, amounting to a total variable cost-to-market of […] EUR/t.

(535) Regarding the variable cost-to-market of Titan, Cimko, W&P and Asamer, in southern Croatia, the Commission has added together the following elements: (i) variable production costs, (ii) costs of freight to reach the Metković terminal (including possible intermediary freights as, for example, for reaching the nearest port from the production plant), (iii) loading and unloading costs, (iv) fees (e.g. port fees, custom fees, etc.), (v) terminal variable costs as specified in the contract with the Port of Ploče, and (vi) cost to reach DDC's customers in southern Croatia from the Metković terminal.

(536) Elements (i)-(iv) were generally provided by Titan, Cimko, W&P and Asamer.

(537) Element (v) is based on the terms of the current contract between Cemex and the Port of Ploče, which will be substantially the same as those in the lease contract with the new lessee. As noted in recital (526), there is a lack of clarity regarding the interpretation of certain clauses of the existing contract, including the clauses concerning the variable costs of handling volumes over […]kt of grey cement. The Commission has therefore analysed two scenarios by applying variable costs of handling volumes of […] EUR/t and of […] EUR/t.

(538) Element (vi) is based on DDC's data and an average estimate of […] EUR per ton. The measure focusses on the distance from the Metković terminal to DDC's customers to make the comparison between the costs of each new lessee that showed an interest in leasing the Metković terminal in response to the Commission's market test and DDC costs on a like-for-like basis. This does not imply, however, that the Commission expects the new lessee to serve the same customers as DDC. On the contrary, each new lessee should be able to compete effectively with the merged entity on a lasting basis in the whole southern Croatia.

Seaborne supplies of grey cement

(539) While Titan from its plant in Fushe Kruje (Albania), Cimko from its plant in the Yumurtalık region in Anatolia (Turkey), and W&P from its plant in Anhovo (Slovenia) indicated that they envisaged using seaborne transport to deliver grey cement to the Metković terminal, each would have significant cost-disadvantages relative to DDC.

(540) Table 11 presents the variable cost-to-market in southern Croatia of Titan and W&P, and the percentage difference with DDC's variable cost-to-market. Table 12 presents the transport costs of Cimko in southern Croatia and the percentage difference with DDC's transport costs.

Table 11 - Variable cost-to-market of DDC, Titan and W&P using seaborne transport

Cost-to-Cost-to- Percentage Percentage market Transport costmarketdifference difference with Cost-to-to reach DDC'swithwith DDC'swith DDC's […]EUR/t market atcustomers in […]EUR/tcost to cost to as MetkovićSouthernas terminalmarket market terminal Croatia variable(EUR (EUR variable cost […]/t) […]/t) cost

Titan (Albania) 50-60 […] […] […]% […] […]%

W&P (Slovenia) 40-50 […] […] […]% […] […]%

The Commission was able to compare only the transport costs for Cimko.

While Titan based its cost estimates on specialised self-discharging cement vessels, Cimko and W&P based their estimates on regular bulk vessels. In the Commission's view, Cimko and W&P's estimates are based on an optimistic and less realistic scenario by relying on less efficient vessel transport and disregarding additional costs. This is because bulk carriers are less efficient than self-discharging vessels to transport cement, as they are more affected by weather conditions and have to rely on on-shore discharging facilities to unload cement, which, at the Metković terminal, [strategic information]. See ID3791, minutes of a conference call with Howe Robinson of 15 February 2017; ID3462, minutes of a conference call with Cemex of 3 February 2017; ID3525, non-confidential version of Port of Ploče's Response to RFI of 14 Feb 2017. ID3500, non-confidential minutes of the meeting of 9 February 2017.

Table 12 - Transport costs of DDC and Cimko using seaborne transport

Transport Percentage Percentage Transport Transport Totalcost to reach difference difference costs with cost with DDC's with DDC's […]EUR/t as […]EUR/t as cost to reachcustomers in transport transport terminal terminal MetkovićSouthern cost (EUR cost (EUR variable costvariable cost Croatia […]/t) […]/t)

Cimko (Turkey) 30 […] […] […]% […] […]%

Table 11 indicates that:

(a) Titan would have a variable cost-to-market disadvantage compared to DDC of […]%, and up to […]% for quantities of grey cement above […]kt.

(b) W&P would have a variable cost-disadvantage relative to DDC in the range of […]%, and up to […]% for quantities of grey cement above […]kt.

Table 12 indicates that Cimko would have a transport cost disadvantage compared to DDC of […]%.

Road-based supplies of grey cement

(543) While Titan from its plants in Kosjerić (Serbia) and Fushe Kruje (Albania), and Asamer from its plant in Lukavac (Bosnia-Herzegovina) indicated that they envisaged using trucks to transport grey cement to the Metković terminal, each would have significant cost-disadvantages relative to DDC.

(544) Table 13 presents the variable cost-to-market in southern Croatia of Titan and Asamer, and the percentage difference with DDC's variable cost-to-market.

Table 13 - Variable cost-to-market of DDC, Titan and Asamer using road-based transport

Cost-to-Cost-to- Percentage Percentage market Transport costmarketdifference difference with Totalto reach DDC'swithwith DDC'swith DDC's […]EUR/t cost to reachcustomers in […]EUR/tcost to cost to as MetkovićSouthernas terminalmarket market terminal Croatia variable(EUR (EUR variable cost […]/t) […]/t) cost

Titan (Serbia) 50-60 […] […] […]% […] […]%

Titan (Albania) 50-60 […] […] […]% […] […]%

Asamer (Bosnia) 50-55 […] […] […]% […] […]%

ID3077, Market Test to competitors, Cimko, question 9. The estimate includes […] EUR/t of custom and port fees as estimated by the Parties. See, Response to RFI of 16 December 2016, Table 12.

The Commission notes that to match the total cost-to-market of DDC, Cimko should have variable production costs of […] EUR/t, which appears highly unlikely.

Titan, however, focussed on seaborne supplies as it did not consider road transport as an effective possibility to resupply the Metković terminal. See ID3500, non-confidential minutes of the meeting of 9 February 2017.

Table 13 indicates that:

(a) Titan would have a variable cost-to-market disadvantage compared to DDC of […]%, and up to […]% for quantities of grey cement above […]kt.

(b) Asamer would have a variable cost-to-market disadvantage compared to DDC of […]%, and up to […]% for quantities of grey cement above […]kt.

One competitor that expressed a view in response to the market test of the Commitments indicated that in case of supplies to the Metković terminal by truck, a new lessee may prefer to deliver grey cement directly to customers and use the terminal as a back-up facility.

(547) The Commission considers that in such scenario:

(a) Titan and Asamer would still have a variable cost-to-market disadvantage compared to DDC of […]% and […]%, respectively, and up to […]% and […]%, respectively, for quantities of grey cement above […]kt; and

(b) substantial direct deliveries of grey cement by Titan and Asamer are unlikely to materialise due to their inability to ensure security of supply to end customers through direct deliveries from their production plants.

Rail-based supplies of grey cement

(548) While Asamer indicated that it envisaged using rail to transport grey cement from its plant in Lukavac (Bosnia-Herzegovina) to the Metković terminal, it would have a significant cost-disadvantage relative to DDC.

(549) Table 14 presents the variable cost-to-market in southern Croatia of Asamer, and the percentage difference with DDC's variable cost-to-market.

Table 14 - Variable cost-to-market of DDC and Asamer using rail transport

Cost-to-Cost-to- Percentage Percentage market Transport costmarketdifference difference Total with to reach DDC'swithwith DDC'swith DDC's cost to […]EUR/t customers in […]EUR/tcost to cost to reach as Southernas terminalmarket market terminal Croatia variable(EUR (EUR variable cost […]/t) […]/t) cost

Asamer (Bosnia) 45-50 […] […] […]% […] […]%

(550) Table 14 indicates that Asamer would have a variable cost-to-market disadvantage compared to DDC of […]%, and up to […]% in case of quantities of grey cement above […]kt.

The variable cost-to-market disadvantages of the potential lessees are not offset by security of supply

(552) The variable cost-to-market disadvantages of the potential lessees are not offset by improved security of supply of sales from the Metković terminal relative to sales from DDC Kakanj.

(553) First, any increase in security of supply arising from the Metković terminal will be lower than the security of supply that each of Cemex and DDC's production assets can currently provide to customers in southern Croatia. Cemex and DDC each have an established customer base, established logistical routes to reach customers and their sales originate from production plants with sizeable production capacity and storage capacity ([…]kt for Cemex; […]kt for DDC Kakanj). By contrast, a new lessee would: (i) not have a customer base; (ii) operate through a terminal with [strategic information]; (iii) have to resupply the terminal from a distant production plant on the basis of newly established logistical routes with associated challenges and uncertainties (see recitals (602) to (606)); and (iv) have to operate at high capacity utilisation rates in order to develop into a viable competitive force that could compete effectively with the merged entity on a lasting basis (see section 9.5.3 for an assessment of the terminal's capacity), which exacerbates the difficulties of ensuring security of supply at the same level as a production plant such as Cemex and DDC's production assets.

(554) Second, while a precise quantification of the value that customers attach to security of supply from the Metković terminal is not possible, the price premium of Cemex's bulk cement sales relative to DDC's sales in southern Croatia is likely to reflect, in part, a security of supply advantage of Cemex Split relative to DDC Kakanj. Cemex's price premium in Southern Croatia over DDC is approximately […] EUR/t, which represents approximately […]% of DDC's variable cost-to-market and is hence lower than the cost disadvantages of each new lessee discussed in recitals (539) to (551).

9.5.2.2. Titan's decision not to pursue the negotiation of the lease

(555) Titan – having had access to the Metković terminal, having entered into exchanges with the Port of Ploče and having had access to the terms of the lease – considers that the lease is not an attractive business proposition for the following main reasons:

(556) First, seaborne supply of grey cement to the Metković terminal would be costly due to the terminal's limitations, with the result that only small self-discharging vessels could be used. There is, however, limited availability of such vessels.

(557) Second, road supply of grey cement to the Metković terminal would be costly and difficult to organise, requiring the use of up to 30-35 trucks, the commercial viability of which is questionable.

(558) Third, using the Metković terminal merely as a back-up facility would not be a viable option due to the extra costs incurred. Titan submits: "Using the Metković terminal only as a back-up facility and not regularly shipping the volumes through it would also not make business sense as it would come with extra cost. Furthermore, even if the Metković terminal was used, there would still be problems with ensuring the required security of supply to large projects which, for instance, may require one truckload of cement every 30 minutes and where just in time delivery may be critical for instance when casting the concrete under a specific schedule."

(559) Fourth, Titan was not willing to commit to pay the annual rental fee against the risks of finding a sufficiently large customer base as a new supplier: "Having fixed volume stated in the contract represents a high risk considering the market conditions and combined with availability of vessels. Should anything occur that would impede the availability of the vessels, Titan would still have to pay the considerable fixed costs for the rent of the terminal. To be acceptable the contract should have lower fixed costs and be more flexible to be adapted to the logistics situation, the consequent constraints of the terminal, and the corresponding market conditions."

(560) In the Commission's view, Titan's assessment confirms the challenges and uncertainty of a new lessee growing into a viable competitive force that could compete effectively with the merged entity on a lasting basis.

9.5.2.3. Asamer's suitability as a new lessee

(561) For the reasons set out in recitals (562) to (578), the Commission concludes that Asamer would not be a suitable new lessee of the Metković terminal.

(562) First, the information provided by Asamer and the Notifying Parties regarding the suitability of Asamer as a new lessee of the Metković terminal was provided after the deadline laid down in Article 19(2) of the Implementing Regulation and was limited.

(563) Second, based on the limited information provided by Asamer and the Notifying Parties after the deadline laid down in Article 19(2) of the Implementing Regulation, the Commission is unable to conclude with the requisite degree of certainty that Asamer is able to develop its grey cement business in the relevant markets as a viable competitive force that could compete effectively with the merged entity on a lasting basis and thus that the Commitments are capable of being implemented effectively within a short period of time.

(564) In the first place, Asamer would have a significant variable cost-to-market disadvantage in comparison with DDC as set out in section 9.5.2.1. This applies regardless of whether Asamer would rely on truck transport or – potentially in the future – on rail transport.

(565) In the second place, Asamer would be unlikely to develop into a viable competitive force that could compete effectively with the merged entity on a lasting basis because of the small scale of the Metković terminal as set out in section 9.5.3.

(566) In the third place, Asamer and the Port of Ploče have agreed on 68 working hours per week of the Metković terminal while the Notifying Parties calculated the capacity of the terminal on the basis of […] working hours per week. Accordingly, Asamer could achieve a maximum capacity of […]% of the capacity claimed by the Notifying Parties for supplies by truck where the unloading time is the bottleneck for the capacity, lowering the nominal capacity from [150-170]kt to [120-150]kt. Assuming that the unloading time is equally the bottleneck for supplies by rail, the nominal capacity would be lowered from […]kt to […]kt.

(567) In the fourth place, Asamer is likely to encounter logistical issues when regularly refilling the terminal by truck. A round trip from Asamer's plant in Lukavac/Bosnia-Herzegovina to the Metković terminal would take around one working day and Asamer would need to make that round trip with 25 to 30 cement trucks per day in peak times of demand. Asamer, however, has limited experience with organising truck transport since only a small part of the logistics for its current cement deliveries is organized by Asamer itself. Asamer relies for the most part of its sales on customer pick-up sales.

(568) In the fifth place, adverse weather conditions in the winter and bad roads have prevented Asamer from selling cement in southern Croatia in the past. Those conditions would equally be a challenge for the 25 to 30 cement trucks per day required to operate the Metković terminal at full capacity.

(569) In the sixth place, the rail connection linking the Metković terminal to the public rail network in Croatia is currently not operational and there is no equipment to unload cement from trains at the terminal. The Notifying Parties have also not presented any information about when and how the necessary investment would be made.

14 March 2017, the Notifying Parties and Asamer informed the Commission of the agreement entered into by Asamer and the Port of Ploče on 13 March 2017 for the lease of the Metković terminal, ID3766 (non-confidential lease agreement with Port of Ploče).

(570) In the seventh place, even if Asamer and the Port of Ploče were to commit to such an investment as suggested by the Notifying Parties, the Commission would have no certainty as to whether such plans would actually take place as they would not be enforceable by the Commission. Moreover, Asamer and the Port of Ploče could jointly decide to waive such an investment at any point in the future.

(571) In the eighth place, Asamer has not presented any further information about how the transport of grey cement to the terminal would be organised and whether there would be sufficient capacity on the public railway network. Asamer has confirmed that it: "would need more information to find the optimal logistics set-up […]."

(572) In the ninth place, the overall condition of the railway network in Bosnia-Herzegovina is poor. The vast majority of the railway network is single track, and in particular the routes Lukavac-Doboj and Zenica-Metković, which represent the longer part of the route from Lukavac to Metković. Moreover, the route Lukavac-Doboj is not electrified. This is also shown by the fact that in 2009 the average speed of freight trains was 32.6km/hr. It appears that no significant improvements have been carried out since then, as the current speed limit for freight trains is 50km/hr.

(573) In the tenth place, while Asamer would require the use of specific silo wagons for the transport of grey cement to the terminal by rail, it is unclear whether a sufficient number of wagons would be available for regular rail transport. According to information provided by the Notifying Parties, the Bosnian railway operator Željeznice Federacije Bosne i Herzegovine (ŽFBH) currently has available only 17 specialised cement rail wagons, although ŽFBH indicates that more could be leased if necessary at "sufficient notice".

(574) In the eleventh place, it appears that one train could only make one shipment per week from Lukavac to Metković. It would take up to two days for a train to make the one-way trip from Lukavac to Metković. Adding extra time for the loading and unloading of the train and for the return journey, this would result in up to 5-6 days for a round-trip. Therefore, it appears that one train could not be used more than once per week.

(575) In the twelfth place, a contractual agreement with three different rail operators (the two Bosnian rail operators and the Croatian rail operator) would be necessary for the supply operations to the Metković terminal by rail transport. This may create further costs and logistical complexities.

(576) In the thirteenth place, in Bosnia-Herzegovina Asamer has shown a lack of aggressiveness in competing with the Parties despite the proximity of its production facilities to DDC's core market (see section 7.7.2.4, 7.8 and, 7.11) and there is no reason to expect that Asamer would compete more aggressively in southern Croatia if it were to enter into a lease for the Metković terminal.

(577) Third, the uncertainties identified in recitals (563) to (576) have not been dispelled by either the agreement entered into by Asamer and the Port of Ploče on 13 March 2017 for the lease of the Metković terminal or Asamer's submission of 2 March 2017 that its supervisory board has agreed to investments for the establishment of the missing rail infrastructure and compressors.

(578) While the agreement and the decision of Asamer's supervisory board indicate Asamer's willingness to lease the Metković terminal, this mere business decision does not provide the requisite degree of certainty that Asamer would also have the ability and the incentives to develop its grey cement business in the relevant markets as a viable competitive force that could compete effectively with the merged entity on a lasting basis. In particular, neither the Notifying Parties nor Asamer have provided any explanation how, notwithstanding the agreement and the decision of Asamer's supervisory board, Asamer intends to overcome the uncertainties described in recitals (560) to (574).

9.5.3. Concerns about the scale of the Commitments

(579) For the reasons set out in recitals (580) to (606)(h), the Commission concludes that the capacity at the Metković terminal will be insufficient for a new lessee to develop into a viable competitive force that could compete effectively with the merged entity on a lasting basis.

(580) First, for the following reasons there is uncertainty regarding the actual capacity of the Metković terminal if refilled by rail. Because of this uncertainty the Commission assumes a capacity of […]kt as presented by the Notifying Parties in the Form RM.

(581) In the first place, the Metković terminal currently has no capacity if refilled by rail transport due to lack of railway infrastructure and unloading facilities.

As ŽFBH recognises, however, the trip could take longer depending on actual loading and unloading times at Lukavac and Metković.

(582) In the second place, the Notifying Parties did not explain in the Form RM what would be the bottlenecks for the capacity of the Metković terminal if refilled by rail.

(583) In the third place, while the Notifying Parties claimed for the first time on 20 February 2017 that the bottleneck would be [strategic information], they still did not provide any specific information on unloading rates or how they had calculated the […]kt capacity. On the same day, the Notifying Parties also claimed that the Metković terminal's capacity could be increased to […]kt if further investments were made to purchase an [strategic information], but again did not provide any information to support their claims.

(584) In the fourth place, there is uncertainty regarding whether and when such investments would be made (see recital (568)).

(585) In the fifth place, there is uncertainty regarding the exact level of investments required.

(586) In the sixth place, Asamer estimates that even if investment in compressors were made, the maximum capacity of the Metković terminal would be only "up to 150kt" and not […]kt as claimed by the Notifying Parties.

(587) Second, due to the seasonality of demand in southern Croatia and southern Bosnia, the effective capacity that the Metković terminal could achieve without incurring capacity constraints would be two-thirds of the capacities claimed by the Notifying Parties, independently of the means of transport through which the Metković terminal would be supplied.

(588) In the first place, in order for a new lessee to develop into a viable competitive force that could compete effectively with the merged entity on a lasting basis, it would have to be able to serve customers also in periods of peak demand.

(589) In the second place, due to demand seasonality, and assuming that the Metković terminal faces a demand equal to its maximum capacity, the terminal would be capacity constrained during […] weeks of the year, in particular during the winter season.

620 ID3509, reply to request for information of 15 February. The Parties provided the unloading rates for seaborne and road-based deliveries, which were at the basis of their calculation of the terminal's capacity, but did not provide this information for rail-based deliveries.

621 ID3509, reply to request for information of 15 February 2017, question 3.

622 The Parties provided several figures regarding the investments. In the Form RM (ID2725), the Parties mention EUR […] as the cost to repair the internal railway and the facilities at the terminal (see Form RM, footnote 13). In the later reply to request for information of 15 February 2017 (ID3509), the Parties claim that the investment required to repair railway connection and unloading facilities at the terminal would amount to […] EUR, and an additional EUR […] to purchase a second compressor.

623 ID3739, non-confidential e-mail of Asamer dated 9 March 2017. If those 150kt are used in the calculations, the capacity of the terminal would still be insufficient in order for Asamer to develop into a viable competitive force that could compete effectively with the merged entity on a lasting basis, as is apparent from the calculations made for the supply of the terminal by trucks (for which the Parties estimated the capacity of the terminal to be 156kt). See recitals (589) to (601).

624 [DDC's market insight]. See Figure 22.

625 The assumption is consistent with the Parties' comparisons of the terminal's capacity with the overall demand in Dalmatia or with DDC's sales (ID2725, Form RM, paragraphs 39 and 47). It is also consistent with the Parties' computation of the maximum capacity of the terminal, which implies a constant sales throughput throughout the year (ID2725, Form RM, footnotes 6-7).

626 The Commission computed the combined seasonality of the demand in Dalmatia and southern Bosnia-Herzegovina (the two closest regions to the Metković terminal) by aggregating the Parties' weekly sales in 2015 and 2016. The Parties' sales are a good proxy of demand because they represent the vast

(590) In the third place, the Metković terminal could only serve a demand of on average […]% (ranging between […]% and […]% based on the seasonality patterns of [year], respectively) of its maximum capacity without incurring capacity constraints in any week of the year. This effectively reduces the available capacity of the Metković terminal by one-third.

(591) Table 15 thus presents the terminal's nominal maximum capacity by means of transport as provided by the Parties and the corresponding seasonality-adjusted maximum capacity. The effective available capacity of the Metković terminal would thus be [120-140]kt if supplied by sea, [100-120]kt if supplied by road, and [60-80]kt if supplied by rail.

Table 15 - Terminal's nominal capacity and seasonality-adjusted capacity by means of transport.

Resupply by sea (kt) Resupply by road (kt) Resupply by rail (kt)

Nominal annual [180-200] [150-170] [100-120] capacity

Seasonality-adjusted [120-140] [100-120] [60-80] capacity

Source: Commission's computation based on data provided by the Parties

(592) Third, due to the vicinity of the Metković terminal to Bosnia-Herzegovina and Montenegro, it is unlikely that a new lessee would sell the entire capacity of the terminal to customers located in southern Croatia. Rather, sales from the Metković terminal are likely to be split among the neighbouring regions. This is for the following reasons.

(593) In the first place, the current catchment area around the Metković terminal not only covers part of Dalmatia but also a substantial part of Bosnia-Herzegovina and part of Montenegro. According to the Notifying Parties' computation, 90% of the current sales from the terminal are delivered within a […] road km distance. Figure 29 provides a representation of the current catchment area around the Metković terminal.

majority of the total sales in the two regions. The Commission then distributed the maximum capacity of the terminal across the seasonality patterns of 2015 and 2016. The possibility to reduce the number of weeks in which the terminal is capacity constrained by an optimisation of logistics appears to be limited due to reduced storage capacity at the customers' premises (between 60 and 100 tonnes). Responses to request for information of 14 February 2017, question 15, ID3411, and 3481.

(594) In the second place, based on the current catchment area of the Metković terminal, the Commission has assumed that the catchment area of the Metković terminal for a new lessee would be between 150 and 200 road km around the Metković terminal. The Commission then computed, in Table 16 and Table 17, the demand for the assumed catchment area around the Metković terminal. The tables indicate that:

(a) within a catchment area of 150 road km around the Metković terminal, [50-60]% of demand is from customers located in Croatia, [30-40]% in Bosnia-Herzegovina and [10-20]% in Montenegro.

(b) within a catchment area of 200 road km around the Metković terminal [30-40]% of the demand is from customers located in Croatia, [50-60]% in Bosnia-Herzegovina and [10-20] in Montenegro.

Table 16 - Cemex's and DDC's sales within a catchment area of 150 road km around the Metković terminal

Cemex's sales (t) DDC's sales (t) Total (t) Percentage

Bosnia […] […] […] […]%

Croatia […] […] […] […]%

Montenegro […] […] […]%

Source: Commission's computation based on data provided by the Parties

Table 17 - Cemex's and DDC's sales within a catchment area of 200 road km around the Metković terminal

Cemex's sales (t) DDC's sales (t) Total (t) Percentage

Bosnia […] […] […] […]%

Croatia […] […] […] […]%

Montenegro […] […] […]%

Source: Commission's computation based on data provided by the Parties

(595) In the third place, in order to assess whether the capacity of the Metković terminal allocated to Croatia would be sufficient to cover the current sales of DDC in southern Croatia, the Commission has applied the proportion of demand originating from Croatia (Table 16 and Table 17) to the seasonality-adjusted capacity of the Metković terminal (Table 15). As indicated in Table 18, the capacity allocated to Croatia would be insufficient to replicate the current sales of DDC in southern Croatia.

632 The Commission notes that if a new lessee's catchment area around the Metković terminal was smaller (up to 100 road km), it would not be able to capture a significant part of the demand in southern Croatia currently captured by DDC and hence it would not be able to replicate DDC's sales in southern Croatia.

633 The Commission approximates the demand in the regions of Croatia, Bosnia-Herzegovina and Montenegro within the catchment area of the Metković terminal by aggregating Cemex and DDC's sales.

634 The increase in the Bosnian demand from the 150 road km to the 200 road km catchment areas is likely due to the inclusion of the demand centre Sarajevo, Bosnia-Herzegovina, in the 200 road km catchment area around the Metković terminal.

Table 18 - Current capacity of the Metković terminal allocated to Croatia and net of DDC's sales.

Metković Catchment Area

150 road km 200 road km

[…] […]DDC sales in Dalmatia

[…]% […]%% demand in Croatia

Resupply by vessels

Seasonality-adjusted capacity (kt) [110-130] [110-130]

[…] […]Capacity allocated to Croatia (kt)

[…] […]Capacity net of DDC's sales (kt)

Resupply by trucks

Seasonality-adjusted capacity (kt) [100-120] [100-120]

[…] […]Capacity allocated to Croatia (kt)

[…] […]Capacity net of DDC's sales (kt)

Resupply by rail

Seasonality-adjusted capacity (kt) [60-80] [60-80]

[…] […]Capacity allocated to Croatia (kt)

[…] […]Capacity net of DDC's sales (kt) Source: Commission's computation based on data provided by the Parties

(596) In the fourth place, even assuming that a new lessee could supplement its sales in the Bosnian part of the catchment area of the Metković terminal with direct deliveries from its production plants, security of supplies issues would limit the amount of such direct sales as explained in recital (546). In case of some direct deliveries in the Bosnian part of the catchment area of the Metković terminal, the Commission considers that the split of demand in the catchment area of 150 road km catchment area would still be relevant.

(597) Fourth, not only would the capacity allocated to Croatia be insufficient to replicate the current sales of DDC in southern Croatia, a new lessee would also have insufficient spare capacity to expand its sales in the event of a post-Transaction price increase by the merged entity.

(598) In order to estimate such additional spare capacity, the Commission has used two proxies.

(599) In the first place, it has computed the level of spare capacity that DDC currently has at its plant in Kakanj, which amounted in 2015 to […]kt. Since that spare capacity can be allocated anywhere in the 250km circular catchment area around DDC's plant, it thus represents the current constraint that DDC exercises on Cemex in southern Croatia.

(600) In the second place, it has computed the proportion of DDC's sales within the Metković terminal's catchment area multiplied by the spare capacity at DDC's

635 In addition, as explained in recital (414), grey cement prices in southern Bosnia-Herzegovina are higher than in southern Croatia. This would give even more incentives to a new lessee to serve southern Bosnia-Herzegovina from the Metković terminal.

636 In general, available spare capacity acts as a constraint to the competitors' prices in that it allows one company to expand its sales in reaction to price increases of competitors. Available spare capacity therefore disciplines the pricing decisions of the competitors reducing their incentives to increase prices.

637 Based on data provided by DDC. ID405-131 M.7878_Form CO_Annex 5.4.III.-RFI 20160706_Response to QP6_DDC.XLSX.

plant ([…]kt). This approach is more conservative because it assumes that spare capacity is allocated proportionally to the sales in a given region. The share of DDC's sales within the Metković terminal's catchment area are […]% and […]%, respectively for the 150 and 200 road km catchment areas of the Metković terminal. Therefore, under the more conservative proxy the required spare capacity at the Metković terminal would be in the range of [20-30]-[60-70]kt.

(601) Based on the two proxies, the Commission considers that the required spare capacity at the Metković terminal should be in the range between 29kt and 100kt. Table 19 presents the Metković terminal's capacity net of DDC's sales (as in Table 18) and net of the required spare capacity. Table 19 indicates that the terminal would have a deficit of capacity in the range of [30-40]kt-[140-150]kt relative to the amount required to effectively constrain the merged entity.

Table 19 - Metković terminal's capacity net of DDC's sales and required spare capacity by mean of transport

Metković catchment area

150 km 200 km

Resupply by vessels

[…] […]Capacity net of DDC's sales (kt)

[…] […]Capacity net of DDC's sales and required spare capacity (29kt/100kt) (kt)

Resupply by trucks

[…] […]Capacity net of DDC's sales (kt)

[…] […]Capacity net of DDC's sales and required spare capacity (29kt/100kt) (kt)

Resupply by rail

[…] […]Capacity net of DDC's sales (kt)

[…] […]Capacity net of DDC's sales and required spare capacity (29kt/100kt) (kt) Source: Commission's computation

(602) Fifth, the clause in the Commitments concerning the back-up facility of grey cement at Cemex's Split plant would not remedy the concerns regarding the insufficient capacity at the Metković terminal. In the first place, relying on the merged entity for such additional capacity would put at risk the independence of a new lessee's business operations. In the second place, the back-up facility at Cemex's Split plant would be unsuitable to remedy short-term shortages of supply as any volumes could be obtained by the new lessee only after it has given [content of internal documents].

(603) Sixth, logistic challenges in the supply of the Metković terminal may further reduce its effective capacity.

(604) In the first place, regarding the supply of the Metković terminal by sea, the effective annual capacity of the terminal is likely to be [100-120]kt or lower rather than [180-200]kt. This is for the following reasons:

(a) In order for the terminal to achieve a capacity of [180-200]kt and for a new lessee to compete effectively with the Notifying Parties, that lessee would have to use self-discharging vessels, which are more efficient in terms of costs, time,

(638) The Commission notes, however, that while […]kt spare capacity would give no uncertainty regarding the ability of a new lessee to effectively constrain the merged entity, lower amounts of spare capacity would imply a higher degree of uncertainty in relation to its ability to constrain the merged entity's pricing.

(639) reliability and cleanliness and thus preferable over standard bulk carriers. There is, however, a limited availability of such vessels able to sail the river connecting the terminal to the sea due to limitations in terms of draft, height

(640) above water and manoeuvrability. Only two suitable vessels, MV Jadro and MV East Coast currently available for the supply of the Metković terminal, thereby, while the Parties' calculations are based on the usage of three vessels, reducing the effective capacity of the Metković terminal by one third.

(b) The Parties' calculations are based on the usage of three vessels on a full-time basis. The only two suitable vessels are, however, already partially used by Cemex and LafargeHolcim, thereby limiting their joint availability to a total of 104 hours per week or one additional trip per week, and the annual effective capacity of the Metković terminal to […]kt (assuming operation of the vessel of 1.25kt in 52 weeks per year).

(c) The use by a new lessee of standard bulk carriers would reduce the effective capacity of the terminal likewise below [180-200]kt because the discharging equipment of the Port of Ploče has not been used since 1996, is outdated and is unprotected against rain and wind and thus cannot be used during bad weather conditions.

(639) ID3791, minutes of a conference call with Howe Robinson of 15 February 2017; ID3462, minutes of a conference call with Cemex of 3 February 2017 and ID3500, non-confidential minutes of the meeting of 9 February 2017. Similarly, the Port of Ploče bases its capacity calculations on the assumption that self-discharging vessels are used without considering the use of standard bulk carriers, ID3140, non-conf version of Port of Ploče's Response to RFI of 2 Feb 2017. LafargeHolcim submits: "In addition, there are technical limitations since cement carrying ships need to be self-discharging/pneumatic to be suitable for the Metković terminal.", ID3437, non-confidential minutes of a conference call of 8 February 2017.

(640) ID3172, LafargeHolcim, reply to Market Test to Competitors, Q 26: "The size of the vessels that can enter Neretva river is the biggest limitation; the second biggest is being able to find available vessels meeting this criteria which could be contracted." Titan submits: "The depth of the river (4m), its bends and the presence of a bridge (14m beam height) only allows the use of vessels with very specific characteristics. Using these vessels implies generally higher freight costs for any future lessee and issues in terms of availability of such vessels.", ID3500, non-confidential minutes of the meeting of 9 February 2017.

(641) LafargeHolcim submits in this context "the biggest limitation on the usage of Metković terminal from our point of view is the fact that only vessels of up to 1.2 kt capacity can be used, and these are not readily available on market. To our knowledge there are two – Jadro and East Coast – but these are both already contracted by third parties." And "the size of vessel that can enter the Neretva river is in our view the biggest limitation and the main factor when calculating distance from which cement supplies to the Metković terminal can be cost competitive. The second challenge is to find available vessels on the market which you could contract to serve your needs. We are aware of only two such vessels present on the market: East Coast and Jadro.", LafargeHolcim reply to Market Test to Competitors, Q5.1., LafargeHolcim reply to Market Test to Competitors, Q4 (ID3172), The availability of only two suitable vessels was confirmed by Titan, ID3500, non-confidential minutes of the meeting of 9 February 2017. This is confirmed also by Howe Robinson submitting further, that the vessels Eastcoast and Jadro are old which might force it to stop operations in case regulations change; ID3891, minutes of a conference call with Howe Robinson of 15 February 2017.

(642) ID3481, Parties' consolidated reply to request for information of 14 February 2017, Q11 and 12.

(643) ID3481, Parties' consolidated reply to request for information of 14 February 2017, 12.

(644) ID3791, minutes of a conference call with Howe Robinson of 15 February 2017; ID3462, minutes of a conference call with Cemex of 3 February 2017; ID3500, non-confidential minutes of the meeting of 9 February 2017.

(d) Using larger vessels of up to 5kt capacity (to the extent that such vessels would be available which is questionable as set out in point (a)) and loading them only up to 1.25kt would lead to higher freight rates.

(e) In general, weather conditions are a factor limiting the capacity of the Metković terminal in case of seaborne supplies.

(605) In the second place, regarding the supply of the Metković terminal by road, the actual capacity of the terminal is likely to be lower than [150-170]kt. This is because, in order for the terminal to achieve a capacity of [150-170]kt, this would require an average of 25-30 trucks per day transporting grey cement from the production plant of a new lessee to the Metković terminal (see recital (564)). Because of the complexity of truck-based supplies, this is, however, unlikely to occur:

(a) "Supplying the Metković terminal by road transport would be quite complicated. Road transport from Kosjeric to Metković would need to cover 400km and would take about 15h one way as the trucks can on average cover only 30-40km/h and there are time-consuming customs formalities at 2 border crossing points. The supply would require the use of a big fleet of trucks (30-35 around the clock)."

(b) "In addition, the option of loading the terminal by trucks makes no economic sense due to distances and transportation costs."

(606) In the third place, regarding the supply of the terminal by rail, the actual capacity of the Metković terminal may be lower than […]kt due to the following reasons:

(a) the Notifying Parties have provided limited information regarding the actual capacity of the terminal and have not explained the bottlenecks in the supply of the terminal until 20 February 2017 as set out in recital (582);

(b) supply by rail of the Metković terminal is currently unfeasible due to the absence of rail infrastructure and specialized unloading facilities at the terminal as set out in recital (569);

(c) the amount of investment required to repair the rail infrastructure and the unloading facilities is unclear (it would appear to be up to EUR […]) and there is uncertainty whether such investments would be carried out as set out in recitals (569)-(570);

(d) the overall condition of the railway network in Bosnia-Herzegovina is poor, as set out in recital (572);

(e) there may be an insufficient number of rail wagons available to carry bulk grey cement, as set out in recital (573);

(f) one train may take 5-6 days for a round-trip from Asamer's Lukavac plant to the Metković terminal, as set out in recital (574);

(g) considering points (e) and (f) above, and that a specialized train wagon carries on average 50 tons, the maximum annual capacity of the terminal would be 46.8kt; and

(h) a contractual agreement with three different rail operators may entail delays and additional costs, as set out in recital (575).

9.5.4. Concerns regarding the implementation modalities of the Commitments

(607) The Commission concludes that the Commitments suffer from a number of defects as regards their implementation modalities:

(608) First, there is no safeguard clause in the Commitments in case the Port of Ploče does not agree to enter into a lease agreement with a new lessee.

(609) Second, there is no clause establishing a time period for the Notifying Parties to find a suitable new lessee.

(610) Third, there is no clause safeguarding the termination and transfer of the lease by Cemex if the Notifying Parties do not find a suitable new lessee.

(611) Fourth, there is no clause stipulating that the Notifying Parties will be deemed not to have complied with the Commitments if no suitable new lessee has been approved by the Commission within a specified deadline.

10. CONCLUSION AND COMPATIBILITY WITH THE INTERNAL MARKET

(612) The Commission finds that the notified concentration will significantly impede effective competition in a substantial part of the internal market within the meaning of Article 2 of the Merger Regulation through non-coordinated effects, which could amount in particular to the creation of a dominant position, in grey cement markets alternatively defined as (i) a 250km circular catchment area around Cemex's Split plant or (ii) a modified 250km catchment area around Cemex's Split plant.

(613) Therefore, the Commission concludes that the notified concentration is incompatible with the internal market and the functioning of the EEA Agreement.

649 One train per week with 18 wagons, 52 weeks per year, see also recital (573).

650 There is only a Memorandum of Understanding signed between the Parties, Titan and the Port of Ploče which, however, is not legally binding. Titan has already decided that despite the Memorandum of Understanding it will not pursue the lease further.

HAS ADOPTED THIS DECISION:

Article 1

The notified operation whereby HeidelbergCement AG and Schwenk Zement KG would acquire joint control of Cemex Hungária Építőanyagok Kft and Cemex Hrvatska dd within the meaning of Article 3(1)(b) of the Regulation (EC) No 139/2004 is hereby declared incompatible with the internal market and the functioning of the EEA Agreement.

Article 2

This Decision is addressed to:

HeidelbergCement AG Berliner Str. 6 69120 Heidelberg Germany

Schwenk Zement KG Hindenburgring 15 89077 Ulm Germany

Done at Brussels, 5.4.2017

For the Commission

(Signed) Margrethe VESTAGER Member of the Commission

Case M.7878 – HeidelbergCement/Schwenk/Cemex Hungary/Cemex Croatia

ANNEX: CAPACITY SHARES BASED ON CUSTOMER LOCATION

(1) The Commission computed, for each customer location served by the Parties, a location specific measure of post-merger concentration of potential suppliers.

(2) For that computation, the Commission used the Parties' data for the sales of cement produced in Cemex' Split plant and DDC's Kakanj plant which included information on (i) the customer identity and location, (ii) the product sold (e.g. bulk/bagged cement, type of cement, etc.) (iii) revenues and costs, and (iv) the distances (measured in truck road kilometers and hours) from the customer location to plant or terminal from which the cement was delivered/picked up, as well as the distances from the customer location to plants and terminals of competitors. In addition, the Commission collected data from the competitors, in particular on their plants and terminals' capacity and on their sales radii, making sure that the methodology applied was consistent across competitors.

(3) The Commission used the data of the Parties and the data of the competitors to compute capacity shares for each of the Parties' customers' locations. That is, for a given customer location, the capacities of all the plants and terminals for which their 90% sales road km radius reached that customer location were counted in the computation of the capacity shares. The definition of a given percentage of sales (i.e. 90%) to compute the plants'/terminals' radii is required to eliminate possible outlier sales.

(4) The methodology is similar to the methodology adopted by the Parties in their elaboration of heatmaps (see Figure 15 to Figure 17 of the Decision). However, the Commission's computation has the following advantages: First, having gathered information from the Parties' competitors, it does not require to make assumption of the 90% sales radii (e.g. using the same radius for each plant), and rather uses the actual radii for each plant and terminal. Second, it uses radii measured in truck road distance km, which, contrary to geodesic distances, are able to take into account the road infrastructure present around each plant/terminal. Third, focusing on the actual customers' locations excludes areas with possibly no demand for cement or where the Parties are not present.

(5) The Commission considers that using plant/terminal-specific 90% sales road km radii, as opposed to the same radius for all plants and terminals as in the Parties' analysis, allows to approximate key factors driving competition in the cement industry such as transport costs and variable costs of production. Such factors differ across plants and terminals, and hence the plant/terminal-specific approach appears to be more precise.

(6) The Commission acknowledges, however, that the 90% sales road km radii are likely to be affected also by the distribution of the demand over the area around the

1 ID1096 Cemex's Reply to RFI of 28 October 2016, and attachments, and ID1099 DDC's Reply to RFI of 28 October 2016, and attachments.

1

(7) The list of plants and terminals of the Parties and of the competitors that are included in the analysis is presented in Table 1.

Table 1: Plants and terminals included in the computation of the capacity shares based on customers location

Company Plants/Terminals

Cemex Croatia Plants: Split

Terminals: Bakar, Bar, Blatna, Bumbarevobrdo, Banja Luka, Kraljevo, Ljubuski, Maksimir, Metkovic, Nis, Podgorica, Tomislavgrad, Varazdin

Heidelberg (DDC, Plants: Kakanj, Beremend (DDC), Trieste (Italcementi) Italcementi)

Terminals: Banja Luka, Sarajevo (DDC)

Lafarge Holcim Plants: Koromacno

Terminals: Zadar, Jastrebarsko

Asamer Plants: Lukavac

Cimsa Terminals: Trieste

Nexe Plants: Nasice

Titan Plants: Kosjeric, Aneta*, Eca*

W&P Plants: Anhovo

Cementizillo Plants: Fanna*

Colacem Plants: Shengjin*

Seament Plants: Elbasan*, Fushe Kruje*

CRH Plants: Plant in Serbia*

(8) The Commission adjusted the capacity figures of Cemex's Split plant and DDC's Kakanj plant to reflect that part of the capacity of these plants is currently being used to supply cement to their respective terminals. This was necessary to avoid that in the computation of the capacity shares based on customers' locations, the capacity of

2 There were no data available for the plants and terminals marked with a '*'. For these plants and terminals, the Commission used a 90% sales road km radius equal to the average of the plants' and terminals' radii for which data were available. Moreover, for the capacity of these plants and terminals the Commission used the Parties' assumptions included in Table 1 the Parties' Reply to RFI of 20 September 2016 ID809.

2

(9) The Commission adjusted the capacity figures of Cemex's Split plant and DDC's Kakanj plant to reflect that part of the capacity of these plants is currently being used to supply cement to their respective terminals. This was necessary to avoid that in the computation of the capacity shares based on customers' locations, the capacity of

3 The Commission's based its adjustment on the sales made through the terminals, and not on the effective capacities of the terminals, to account for the fact that certain terminals may be relatively inactive and used only for storage purposes, not requiring constant re-supply. At the same time, the terminals' capacities were not adjusted by their percentage sales to take into account that these capacities may in principle be used to supply to end customers.

3