Commission, April 18, 2013, No M.6843

Dear Sir/Madam,

Subject:  Case No COMP/M.6843 - Siemens/ Invensys Rail Commission decision pursuant to Article 6(1)(b) of Council Regulation No 139/20041

(1)   On 12 March 2013, the European Commission received a notification of a proposed concentration pursuant to Article 4 of Council Regulation (EC) No 139/2004 by which Siemens AG ("Siemens", Germany) acquires sole control over Invensys Rail, a division of the company Invensys plc. ("Invensys Rail", United Kingdom) by way of acquisition of shares.2 Siemens and Invensys Rail are collectively referred to as "Parties".

1.        THE PARTIES AND THE CONCENTRATION

(2)  Siemens is a German stock corporation which offers a wide range of products and services to the customers in four business sectors: Energy, Healthcare, Industry, Infrastructure and Cities. The latter includes the Mobility and Logistics division which encompasses Siemens global railway signalling business.

(3)  Invensys Rail is a global provider of railway signalling projects and products.

(4)  The proposed transaction involves an acquisition of sole control by Siemens over Invensys Rail within the meaning of Article 3(1)(b) of the Merger Regulation since Siemens will directly and indirectly acquire shares in various entities, which together constitute Invensys Rail. Invensys Rail will be integrated into Siemens' Business Unit Rail Automation.

2.     EU DIMENSION

(5)    The undertakings concerned have a combined aggregate world-wide turnover of more than EUR 5 000 million3 (Siemens: EUR 78.3 billion, Invensys Rail: EUR […]). Each of them has a EU-wide turnover in excess of EUR 250 million (Siemens: EUR […], Invensys Rail: EUR […]), but they do not achieve more  than two-thirds of their aggregate EU-wide turnover within one and the same Member State. The notified operation therefore has an EU dimension.

3.          COMPETITIVE ASSESSMENT

3.1.         Relevant market

3.1.1.  Relevant product market

(6) The activities of the Parties overlap in the field of railway signalling projects and of railway signalling products.

3.1.1.1.Railway signalling projects

(7)    Railway signalling systems are designed to control the railway traffic and to guarantee the safe operation of trains and railway network, by avoiding train collisions. They can also manage railway traffic and improve rail network efficiency. Railway signalling projects are comprehensive solutions involving: project specific engineering, development and project management, procurement of the necessary equipment, installation, testing and, in most cases, maintenance.

(8)   Railway signalling projects include four key systems.

-        Interlocking systems, which prevent trains from colliding and consist of signals (trackside and/or display on monitor in driver's cabin), trackside sensors (detect if a given sector of railway network is occupied by a train or vacant), point machines (move the rail at junction) and electronic interlockings (process the information generated by trackside sensors and operate the point machines and sensors).

-        Automatic Train Protection systems ("ATPs"), which protect each individual train by assisting the driver and intervene in case of human failure.  ATPs  for  mainline  can  be either  conventional,  developed  at a national level, or conform with the European standard - European Train Control System ("ETCS").

-        Operations and control systems, which operate interlockings, integrate information generated by interlockings and ATP systems and can also manage passenger information, fleet and crew.

-        Level crossing systems, which include components predicting train arrival, barriers and signals for intersections of railway line with a road or path.

(9)   The Commission has previously investigated the market for railway signalling4 and analysed whether this market can be further subdivided according to (i) the rail network type, (ii) the technology applied and (iii) the size of the project.5 However, the exact definition of the relevant product market was ultimately left open.

(a)     Segmentation based on rail network type

·     Railway signalling projects for mainline and mass transit

(10)  The notifying party considers that projects for mainline and projects for mass transit may belong to two distinct product markets because the former are more complex technologically, more expensive, awarded in more burdensome tender procedures, and more challenging for suppliers because they have to be compatible with a wider network. The customers for these two network types are also different (national or regional railway operators for mainline and local operators for mass transit). On the other hand, the notifying party admits that railway signalling projects for mainline and mass transit are based on the same core technology, that there is supply-side substitutability between them, and that the main players in the EEA are essentially active in both segments.

(11)  The market investigation confirmed that the technology required for railway signalling projects for mainline is more sophisticated than the  technology  for mass transit and that, therefore, the projects for mainline are more expensive.6 However, contractual terms in these two segments depend rather on the customer than on the rail network type.7 The market investigation did not establish clearly whether the services offered in these segments are similar or different.8

·     Railway signalling projects for mass transit: metro  and light rail

(12)  The notifying party submits that projects for mass transit may be further segmented into projects for metro and light rail, because of differences in safety requirements and in price.

(13)  The market investigation confirmed that signalling projects for light rail are simpler in technical terms, have lower security requirements (because light rail relies mostly on the driver and signalling systems serve merely as backup and supervision) and are therefore cheaper than signalling projects for metro.9

(b)    Segmentation based on the technology applied

(14)  In 2007, the Commission noted that out of the two main technologies applied in signalling systems, i.e. the non-electronic technology and the computer-based technology, the former is rather outdated.10 The notifying party shares this view and considers that this distinction is only relevant for interlocking systems and that virtually all signalling infrastructures that are being installed in the EEA are based on electronic interlocking.

(15)  However, according to the majority of market participants, non-electronic railway signalling projects are still in operation, in particular on low traffic railway lines and where customers are subject to economic constraints. Market participants estimate that non-electronic projects will remain in use for the next 10 to 50 years.

(c)    Segmentation based on the size of project

(16)   The market investigation in Alstom UK / Balfour Beatty indicated that railway signalling projects could be segmented into: small (value below EUR 8 million), medium-size (value between EUR 8 million and EUR 75 million) and large (value above EUR 75 million) projects. 12

(17)   According to the notifying party, such a segmentation would not be appropriate, since competitive conditions for all projects are essentially the same, all or most major competitors participate in most tenders, and it would not be reasonably possible to draw a meaningful dividing line between various railway signalling projects.

(18) Some market participants agree that a distinction between railway signalling projects on the basis of their size is justified and reflected in different prices and technical complexity required for small, medium and large projects. The market investigation also indicated that local suppliers of railway signalling tend to favour smaller projects and participate in tenders for large or medium projects mostly as sub-contractors or sub-suppliers of the bigger players because they do not have the required financial and technical background.13   On the other hand, customer expectations remain the same regardless of the size of the project.14 Market participants provided very different dividing lines between projects, ranging from below EUR 100.000 to below EUR 10 million for small projects and from above EUR 500.000 to above EUR 100 million for large projects.

(19)  However, the exact market delineation of the relevant product market(s) can be left open as the notified operation does not raise serious doubts as to its compatibility with the internal market under any plausible market definition.

3.1.1.2.     Railway signalling products

(20) Railway signalling products are signalling components used in railway signalling projects. The two particular products analysed in the previous Commission decision were: interlocking equipment and point machines.15 However, the exact market definition was ultimately left open.

(21) The Parties' activities only overlap in the following products :

(a) Track circuits, which are devices indicating whether a given sector of railway track (a block) is vacant or occupied by a train.

(b) Point machines, which are devices used to move a set of rails to allow a train to pass from one track to another.

(c)  Track signals, which are a set of colour lights or mechanical arms installed next to or above the track, showing the train driver to pass, stop or be cautious.

(d)   Conventional on-board units, which consist of an on-board computer with soft- and hard-ware that interprets signalling information and compares it to the trains' current performance in order to determine if any action is required.

(e)   Relays, which are devices that respond to a small current or  voltage change by activating switches or other devices in an electric circuit, and are used in interlockings and level crossings.

(f)  Balises (also referred to as track transponders or a beacons), which are devices installed between the rails to allow for signalling information to be sent directly to the train passing above it.

(22)   For the vast majority of customers and competitors conventional and ETCS balises are not interchangeable16 and for competitors switching from the production of conventional balises to ETCS balises would require significant investment and/or lead time.17 Therefore the market investigation suggests that ETCS balises and conventional balises belong to different markets.

(23)   However, the exact product market definition with regard to balises and to all other railway signalling products can be left open, since the proposed transaction does not raise serious doubts as to its compatibility with the internal market under any plausible market definition.

3.1.2.    Relevant geographic market

3.1.2.1.Railway signalling projects

(24) The notifying party claims that price levels for railway signalling projects are essentially similar across the EEA because of open tenders, of the fact that safety standards are set at a pan-European level and of the increasing rollout of the ETCS. Thus, according to the notifying party, the market for railway signalling projects is at least EEA-wide.

(25)   In previous cases, the Commission acknowledged arguments according to which the markets for railway signalling projects are national in scope18 but also noted the results of the market investigation suggesting that the market is at least EEA- wide.19  Ultimately, the geographic market definition was left open.

(26)  During the market investigation, some market participants noted that railway signalling projects are comparable in the EEA or even worldwide, due to similar technologies, comparable safety requirements, progressing standardisation and exchange of information between customers.20 On  the other hand, customer preferences, national technical specifications, national safety standards and national authorisation processes have often been mentioned as barriers to entry.21

(27)  However, the exact geographic market definition can be left open, since the proposed transaction does not raise serious doubts as to its compatibility with the internal market under any plausible market definition.

3.1.2.2.   Railway signalling products

(28)  In a previous decision, the Commission noted that the relevant geographic market for railway signalling products appears to be national.22 However, the exact market definition was ultimately left open.

(29) Most of the participants in the market investigation considered that obtaining national authorisation for the railway signalling products is not easy since the authorisation processes are different in different Member State and can be time consuming.23 Therefore, it cannot be excluded that the market for railway signalling products should be considered national.

(30)  However, the exact geographic market definition can ultimately be left open, since the proposed transaction does not raise serious doubts as to its compatibility with the internal market under any plausible market definition.

3.2.         Competitive assessment

3.2.1.          Railway Signalling Projects

3.2.1.1.     Market Shares of the Parties and of their Main Competitors

(a)   Segmentation based on rail network type

(31)  At an EEA level, the Parties estimate that, based on the cumulated intake orders for the years 2008-2012,24 their market shares and the market shares of their main competitors are the following:

(32)  In all Member States but the ones mentioned in paragraph (33) below, the Parties have no overlapping activities in mainline, mass transit or metro.26

(33)   At a Member State level, the Parties estimate that, based on the cumulated intake orders for the years 2008-2012, their market shares and the market shares of their main competitors are the following in the Member States where they    overlap:

(34)  With respect to the abovementioned table, it should be noted that, although market shares are normally important factors in the assessment, they only provide first indications of market power and other important factors should also be taken into account such as closeness of competition and buyer power.32 In particular, "[e]ven firms with very high market shares may not be in a position, post-merger, to significantly impede effective competition, in particular by acting to an appreciable extent independently of their customers, if the latter possess countervailing buyer power." 33 As will be shown in sections 3.2.1.2 and 3.2.1.4 below, the Parties are not each other's closest competitor and the customers have significant buyer power. Therefore, the notified operation does not raise serious doubts as to its compatibility with the internal market.

(b)   Segmentation based on the technology applied

(35)  If the market(s) were to be further segmented in order to distinguish between electronic and non-electronic signalling systems, the Parties estimate that their market shares for the years 2008-2012 in a segment limited to non-electronic mainline signalling projects in the EEA would be [10-20]% for Siemens and below [0-5]% for Invensys Rail. The Parties’ activities in the area of non- electronic mainline signalling projects in the EEA do not overlap on a Member State level: Siemens is only active in Austria, Finland and Germany whereas Invensys Rail is only active in the UK and in Spain.34

(36)  Since non-electronic signalling systems represent a small proportion of the overall signalling systems projects,35 the market shares for electronic mainline signalling projects are comparable to the market shares for the overall mainline signalling projects. The Parties estimate that their shares at the EEA level for electronic mainline projects would be [10-20]% for Siemens and [5-10]% for Invensys Rail.

(37) At a Member State level, the Parties’ shares for electronic mainline signalling projects in the four Member States where their activities overlap are virtually the same as for mainline signalling projects overall (i.e. electronic and non- electronic altogether):

-  In Germany, Siemens has an estimated share of [40-50]% whereas Invensys Rail has one of [0-5]%.

-  In Finland, Siemens has an estimated share of [10-20]% whereas Invensys Rail has one of [0-5]%.

-   In Spain, Siemens has an estimated share of [0-5]% whereas Invensys Rail has one of [20-30]%.

-  In the United Kingdom, Siemens has an estimated share of [0-5]%36 whereas Invensys Rail has one of [30-40]%.

(c)     Segmentation based on the size of project

(38)  If the market(s) were to be further segmented in order to distinguish between small  projects  (below  EUR  8 million),  medium  projects  (between  EUR     8 million and EUR 75 million) and large projects (over EUR 75 million),37 the Parties' estimated combined market shares at an EEA level would be: [10- 20]% for small mainline projects,38 [20-30]% for medium mainline projects,39 [20-30]% for large mainline projects,40 [10-20]% for small metro projects,41 [30-40]% for medium metro projects,42 and [30-40]% for large metro projects.43

(39)  On the basis of this segmentation at a Member State level, the Parties consider that, taking into account the 2008-2012 period, their activities would only overlap in the following States:

-  Germany for small mainline projects (with a negligible overlap of less than [0-5]%),

- Spain for small mainline projects (with market share estimates of [5- 10]% for Siemens and [0-5]% for Invensys Rail), medium mainline projects (with market share estimates of [0-5]% for Siemens and [20- 30]% for Invensys Rail), small metro projects (with market share estimates of [0-5]% for Siemens and [40-50]% for Invensys Rail), and medium metro projects (with market share estimates of [10-20]% for Siemens and [50-60]% for Invensys Rail);

-  Finland for small mainline projects (with market share estimates of [10- 20]% for Siemens and [0-5]% for Invensys Rail);

-   the United Kingdom for small mainline projects (with a negligible overlap of less than [0-5]%) and for medium metro projects (with market share estimates of [20-30]% for Siemens and [10-20]% for Invensys Rail).

(d)   Assessment

(40)  At an EEA-level, as can be seen from the market share data above, the notified operation would lead to combined market shares below 30% under most relevant market definitions. In all cases, the merged entity will be facing powerful competitors such as Thales, Ansaldo, Bombardier and Alstom. These competitors are, or belong to, multinational groups with strong financial backbones and world-wide activities. Some customers44 and competitors45 also mentioned during the market investigation that there have been or will be new companies entering the market(s) for railway signalling projects, in particular General Electric.

(41) At a Member State level, the notified operation only gives rise to a  few affected markets and, in most cases, the market share increment is small, i.e. below 5%:

-   In Finland, the Parties' activities only overlap in the segment for mainline signalling projects, where neither Siemens nor Invensys Rail has high market shares. However, Siemens has high market shares in the mass transit/metro segments since it won the tender to replace and extend the only existing metro line in Finland. As a consequence, the notified operation would give rise to an affected market if all signalling projects are taken into account. In such a case, however, the market share increment would only be [0-5]%.

-   In Germany, Invensys Rail has negligible activities in mainline signalling projects and no activities in mass transit/metro signalling projects.

-  In Spain, the notified operation would give rise to affected market regardless of the product market definition but, in most cases, the market share increment would be below 5%. In the market(s) for mass transit/metro signalling, where the increments would be above 5%, the notifying party submits that the market share data tends to overestimate Invensys Rail's position since, due to the economic crisis, there have been few new projects and Invensys Rail therefore benefitted from his incumbent position to obtain follow-on works. However, the notifying party argues that Invensys Rail's position is increasingly being challenged by its competitors, […] and that other companies have been chosen  for  the  new  metros  in  Sevilla,  Barcelona  and  Malaga.    As mentioned  in  paragraph  (34)  above  and  as will  be shown  in sections

3.2.1.2 and 3.2.1.4 below, the Parties are not each other's closest competitor and the customers have significant buyer power.

-  In the United Kingdom, the notified operation would only give rise to overlaps and combined market shares in excess of 15% on the markets for overall projects and for mainline projects (with market share increments of [0-5]% or below) and on the market for medium metro signalling projects.

(42)   In the above-mentioned Member States, the Parties will face the same strong competitors as at the EEA level (i.e. Thales, Alstom, Ansaldo or Bombardier). The Parties will also face local competitors such as Mipro in Finland (with a [20- 30]% estimated market share in mainline signalling projects), CAF, Electrans, ENYSE and Indra in Spain, Atkins, Deltarail and Amey in mainline projects in the United Kingdom (with estimated market shares of respectively [10-20]%, [5- 10]% and [0-5]%), and Balfour Beatty in metro projects in the United Kingdom (with an estimated market share of [10-20]%).

3.2.1.2.Closeness of competition

(43)  The Parties are not each other's closest competitor at an EEA or at a Member  State level. Their strategies differ: Siemens focuses on Germany, Austria, Belgium, Denmark, France, Slovakia, the Netherlands, Greece and Latvia whereas Invensys Rail is almost exclusively active in Spain, the United Kingdom and Portugal.46

(44)  In the EEA, over the past five years, Siemens and Invensys Rail rarely participated in the same tenders and were almost never placed first and second in the same tender. This is the case both for tenders in the Member States where the Parties' activities overlap (as demonstrated by the bidding data provided by the Parties and discussed in detail in paragraphs (46) and (47) below) and in the remaining Member States for mainline47   and metro.48  The fact that the Parties are not each other's closest competitor in the EEA has been confirmed by the majority of customers in their replies to the market investigation. 49

(45) At Member State level, the Parties' are not each other's closest competitor either.

(46) In particular, as regards mainline signalling projects, the bidding data in the Member States where the Parties' activities overlap reveals the following:

-  In Spain, [explanation that the Parties did not or not often meet in tenders in that Member State].

-  In Germany, [explanation that the Parties did not or not often meet in tenders in that Member State]

-  In Finland, [explanation that the Parties did not or not often meet in tenders  in that Member State].

-  In the United Kingdom, [explanation that the Parties did not or not often meet in tenders in that Member State].

(47)  As regards metro signalling projects, the bidding data in the Member States where the Parties' activities overlap shows that:

-  In Spain, [explanation that the Parties did not or not often meet in tenders in that Member State].

-  In Germany, [explanation that the Parties did not or not often meet in tenders in that Member State].

-  In Finland, [explanation that the Parties did not or not often meet in tenders  in that Member State].

-  In the United Kingdom, [explanation that the Parties did not or not often meet in tenders in that Member State].

3.2.1.3.Elimination of an Important Competitive Force

(48)  In the EEA, Invensys Rail is not an aggressive competitor outside its "home countries" (i.e. Spain, the United Kingdom and Portugal). As regards mainline signalling projects, over the past five years, Invensys Rail only participated in […] reported tenders in the EEA outside of its "home countries" and only won […] small projects.50 As regards metro signalling projects, over the  past  five years, Invensys Rail only participated in […] reported tenders in the EEA outside of its "home countries" and only won […] projects in Romania and Norway, the value of which was below EUR […] (only one project was above EUR […]).

(49)   At Member State level, Invensys Rail was not an aggressive competitor in Germany and Finland. Indeed, in Germany, Invensys Rail only participated in […] out of […] reported tenders in mainline and in […] in metro. In Finland, Invensys Rail participated in only […] out of […] reported tenders in mainline and in […] in metro.

(50)   Conversely, in the overlap Member States where Invensys Rail had strong positions, i.e. Spain and the United Kingdom, Siemens was not an aggressive competitor either. In Spain, Siemens only participated in […] out of […] reported tenders in mainline and in […] out of […] reported tenders in metro. In the United Kingdom, Siemens only participated in […] out of […] reported tenders in mainline and in […] out of […] reported tenders in metro.

(51)    It follows from the above that the transaction does not eliminate an important competitive force either at EEA level or at Member State level.

3.2.1.4.Buyer power

(52)  The demand on railway signalling markets is highly concentrated, typically with one customer for mainline and one to five customers for mass transit. These customers apply sophisticated tender procedures which grant them significant buyer power. In particular, a majority of customers who responded to the market investigation confirmed that they usually manage to impose lower prices, contracts with terms beneficial for them, and higher quality requirements51, which can however be satisfied by a number of large players throughout the EEA (e.g. Alstom, Thales, Bombardier, and Ansaldo).

(53)  The significant buyer power is confirmed by a large majority of competitors of the  Parties.52   Some  of these  competitors  mention,  in particular,  the customers' ability to impose contracts with beneficial terms for them and threats to choose different suppliers.53

3.2.1.5.Conclusion

(54)   In light of the above, it can be concluded that the notified operation does not raise serious doubts as to its compatibility with the internal market with respect to the railway signalling projects under any plausible market definition.

3.2.2.          Railway signalling products

(55) The markets for the sale of signalling products on a standalone basis are relatively small, since only limited sales take place outside railway signalling projects (non- captive sales), as confirmed also by the market investigation.54 In the EEA, the markets for the sale of standalone signalling products are generally below EUR 50 million with the exception of on-board units where the market value is approximately EUR […]. In Spain and Germany, the markets for the products at issue always amount to less than EUR 15 million.

(56)    Moreover, although there are Member States where one of the Parties may have a strong position in standalone sales of a specific signalling product, in these Member States the other Party has no sales and thus the proposed transaction would not affect competition in that regard.

(57)    The table below provides market shares for the Parties in 2012 to the extent that their activities overlap in the EEA or in specific Member States.

Shares of supply ranges – Signalling products sourced outside projects 2012

(58) The Parties' activities in the sale of signalling products on a standalone basis overlap in the following possible affected markets:

(a)               At the EEA level with regard to:

–    Point machines: the combined market share of the Parties would be, however, in the range of 25-35% with a small increment of less than 5%. There are several other important point machine manufacturers active in the EEA, such as Alstom, Thales, General Electric, Bombardier, Ansaldo, Vossloh, Voest Alpine Engineering and Vialis NMA.

–   Track signals: the combined market share of the Parties would be in the range of 15-25% with a small increment of less than 5%. There are several alternative suppliers across the EEA, such as Alstom, Thales, Bombardier, Pintsch Tiefenbach, Vialis, AzD Praha, SLK and Zelisko.

–   Relays: the combined market share of the Parties would be in the range of 30-50% with an increment of 5-15%. All project suppliers that are active in the EEA, such as Alstom, Thales, Bombardier, Ansaldo, also supply relays for their respective interlockings. In 2012 the Parties' sales of relays did not overlap in any Member State, and thus they have entirely different customers, i.e. the respective railway and metro infrastructure operators in each of these Member States. Therefore, the Parties appear to be geographically distant competitors with different customers.

–  On-board units: the combined market  share of the Parties  would be in the range of 20-35% with a small increment of less than 5%. Other significant manufacturers in the EEA include Alstom, Bombardier, Thales and Ansaldo.

(b)  At Member State level with regard to track circuits in Spain: with sales of approximately EUR […] in 2012 by Invensys Rail, the increment would be minimal ([0-5]%). Furthermore, certification of track circuits in Spain takes place by the monopsonist customer, ADIF, who has so far certified the track circuits of the Parties but also of at least two other important players, Ansaldo and Bombardier.

(59) Furthermore, in all the above mentioned standalone sales of signalling products the respective customers are national/regional monopsonists with significant buyer power, namely the railway infrastructure operators at a national level or the metro operators of specific cities. Some end-customers confirmed that they prefer to rely on one or few product suppliers55 so that the operation of their signalling infrastructure and in particular the training, maintenance and logistics can be organised more efficiently. In these sales the Parties face strong competitors with EEA-wide activity, such as Ansaldo, Alstom, Thales and Bombardier.

(60)   The market investigation also confirmed that none of the abovementioned railway signalling products could be used by the Parties in order to exert market power in the markets for railway signalling projects, e.g. through refusal to sell an indispensable railway signalling product to a competing provider of a railway signalling project. The vast majority of customers (approx. 70%) stated that, when needed, suppliers of railway signalling products sell these products to their competitors.56 The vast majority of competitors also stated that they purchase (approx. 79%) and sell (approx. 57%) railway signalling products on a stand- alone basis from/to their competitors.57

(61)  Most importantly, approximately 93% of customers58 and 80% of competitors59 were not aware of any situation where a supplier refused to supply another with a railway signalling product that was necessary for a given project or supplied such a product at a very high price. Even the limited number of competitors and customers who replied differently referred rather to difficulties which occurred during negotiations for the supply of certain products, but which were ultimately resolved.

(62)    Finally, during the market investigation no substantiated concerns were raised and the vast majority of customers did not identify any negative impact of the proposed transaction on prices, availability and range of railway signalling products.60

(63)   In view of the above, the proposed transaction does not raise serious doubts as to its compatibility with the internal market with respect to the railway signalling products under any plausible market definition.

4.            CONCLUSION

(64)     For the above reasons, the European Commission has decided not to oppose  the notified operation and to declare it compatible with the internal market and with the EEA Agreement. This decision is adopted in application of Article 6(1)(b) of the Merger Regulation.

1   OJ L 24, 29.1.2004, p. 1 ("the Merger Regulation"). With effect from 1 December 2009, the Treaty on the Functioning of the European Union ("TFEU") has introduced certain changes, such as the replacement of "Community" by "Union" and "common market" by "internal market". The terminology of the TFEU will be used throughout this decision.

2    Publication in the Official Journal of the European Union No C 80, 19.03.2013, p.16.

3    Turnover calculated in accordance with Article 5(1) of the Merger Regulation and the Commission Consolidated Jurisdictional Notice (OJ C95, 16.04.2008, p1).

4       Case No. IV/M. 685 – Siemens/Lagardère, Commission Decision of 8 February 1996; Case No. COMP/M.2694 – Metronet/Infraco, Commission Decision of 21 June 2002; Case No. COMP/M.4337 – Thales/Alcatel, Commission Decision of 7 November 2006; Case No.COMP/M.4508 – Alstom UK/Balfour Beatty/JV, Commission Decision of 30 March 2007.

5       Case No. COMP/M.4508, op. cit., paragraph 13.

6       See responses to question 6.1 of the Questionnaire to Competitors and 8.2 of the Questionnaire to Customers.

7       See responses to question 6.2 of the Questionnaire to Competitors.

8       See responses to question 8.3 of the Questionnaire to Customers and to question 6.3 of the Questionnaire to Competitors.

9       See responses to questions 8.1, 8.4 of the Questionnaire to Customers and to question 6.1 of the Questionnaire to Competitors.

10    Case No. COMP/M.4508, op. cit., paragraph 12.

11    See responses to question 9 of the Questionnaire to Customers and to question 7 of the Questionnaire to Competitors.

12    Case No. COMP/M.4508, op. cit., paragraph 12.

13    See responses to question 11 of the Questionnaire to Customers and to question 8.1 of the Questionnaire to Competitors.

14    See response of a customer to question 11 of the Questionnaire to Customers.

15    Case No. COMP/M.4508, op. cit., paragraphs 15-21.

16    See responses to question 16 of the Questionnaire to Customers and responses to question 13 of  the Questionnaire to Competitors.

17    See responses to question 14 of the Questionnaire to Competitors.

18    Case No. COMP/M.4508, op. cit., paragraph 23.

19    Case No. COMP/M.4337, op.cit, paragraph 22.

20    See responses to question 17 of the Questionnaire to Competitors.

21    See responses to question 18 of the Questionnaire to Competitors.

22    Case No. COMP/M.4508, op. cit., paragraph 26.

23    See response to question 51 of the Questionnaire to Customers and response to question 49 of the Questionnaire to Competitors.

24    The Parties' market share estimates are based on order intakes, which measure the value of railway signalling projects awarded to each player in a given year. Order intakes are commonly used by the industry to calculate market shares: UNIFE, the association of European rail industry, uses order intakes in its reports. The five year timeframe takes into account the duration of railway signalling projects which is typically one to five years (see response to question 25 of the Questionnaire to Competitors).

25  Invensys Rail has no activities in light rail. As a consequence, there are no effects of the notified operation on a hypothetical light rail market, whether EEA-wide or at a Member State level.

26  In Denmark and the Netherlands, Siemens is active but Invensys Rail is not – […]. Conversely, in Portugal, Invensys Rail is active but Siemens is not – […]. In Romania and Norway, Siemens is only active in mainline whereas Invensys Rail is only active in metro.

27  Invensys Rail is not active in the mass transit/metro signalling sector in Finland.

28  Invensys Rail is not active in the mass transit/metro signalling sector in Germany.

29   Invensys Rail's activities in Germany are minimal with only EUR […] of intake orders in the last five years relating to mainline signalling projects.

30   Invensys Rail's activities in Germany are minimal with only EUR […] of intake orders in the last five years relating to mainline signalling projects.

31  Siemens' activities in the mainline signalling sector in the United Kingdom are minimal with only EUR […] of intake orders over the last five years.

32    See Commission Guidelines on the assessment of horizontal mergers under the Council Regulation on the control of concentrations between undertakings, OJ C31, 5.02.2004, p. 5, paragraphs 27, 28 and 64.

33    See Commission Guidelines on the assessment of horizontal mergers under the Council Regulation on the control of concentrations between undertakings, OJ C31, 5.02.2004, p. 5, paragraph 64.

34    The Parties are not aware of any non-electronic signalling projects in metro and light rail in the EEA and do not generate any order intake in this regard.

35    Siemens estimates that approximately [10-20]% of the total order intake for mainline projects over the past five years in the EEA related to such non-electronic signalling projects. Siemens’ cumulated order intake generated through non-electronic signalling projects in the EEA amounted to around EUR […] over the past five years. The equivalent figure for Invensys Rail is around  EUR […].

36    Siemens' activities in the mainline signalling sector in the United Kingdom are minimal with only EUR […] of intake orders over the last five years.

37    In paragraphs (38) and (39), the market shares for mass transit are omitted since the market shares for metro signalling already provide the market share under the narrowest possible market definition (Invensys is not active in light rail).

38    Siemens: [10-20]%, Invensys: [0-5]%. According to the Parties' estimates, there are other significant players on this segment such as Thales ([10-20]%) and Alstom ([10-20]%).

39    Siemens: [10-20]%, Invensys: [10-20]%. According to the Parties' estimates, there are other significant players on this segment such as Thales ([10-20]%) and Alstom ([10-20]%).

40    Siemens: [0-5]%, Invensys: [10-20]%. According to the Parties' estimates, there are other significant players on this segment such as Thales ([20-30]%) and Alstom ([20-30]%).

41    Siemens: [10-20]%, Invensys: [5-10]%. According to the Parties' estimates, there are other significant players on this segment such as Thales ([10-20]%), Ansaldo ([10-20]%) and Alstom ([10-20]%).

42    Siemens: [20-30]%, Invensys: [5-10]%. According to the Parties' estimates, there are other significant players on this segment such as Ansaldo ([10-20]%) and Alstom ([10-20]%).

43    Siemens: [30-40]%, Invensys: [0-5]%. According to the Parties' estimates, there are other significant players on this segment such as Ansaldo ([20-30]%) and Bombardier ([30-40]%).

44    See responses to question 34 of the Questionnaire to customers.

45    See responses to question 27 of the Questionnaire to competitors.

46    See e.g., internal documents provided in Annex 17 to the Form CO.

47    For the other EEA States, the bidding data submitted by the notifying party for the years 2008- 2012 in mainline projects is as follows: in Denmark, [explanation that the Parties did not or not often meet in tenders in that Member State]; in the Netherlands, [explanation that the Parties did  not or not often meet in tenders in that Member State]; in Portugal, [explanation that the Parties did not or not often meet in tenders in that Member State]; in Romania [explanation that the Parties did not or not often meet in tenders in that Member State]; in Austria, [explanation that the Parties did not or not often meet in tenders in that Member State]; in Belgium, [explanation that the Parties did not or not often meet in tenders in that Member State]; in Bulgaria, [explanation that the Parties did not or not often meet in tenders in that Member State]; in the Czech Republic, [explanation that the Parties did not or not often meet in tenders in that Member State]; in Estonia, [explanation that the Parties did not or not often meet in tenders in that Member State]; in Hungary, [explanation that  the Parties did not or not often meet in tenders in that Member State]; in Italy, [explanation that the Parties did not or not often meet in tenders in that Member State]; in Lithuania, [explanation that the Parties did not or not often meet in tenders in that Member State]; in Latvia, [explanation that the Parties did not or not often meet in tenders in that Member State]; in Poland, [explanation that the Parties did not or not often meet in tenders in that Member State]; in Sweden, [explanation that the Parties did not or not often meet in tenders in that Member State]; in Slovakia, [explanation  that the Parties did not or not often meet in tenders in that Member State]; in Slovenia,  [explanation that the Parties did not or not often meet in tenders in that Member State]; in Norway,[explanation that the Parties did not or not often meet in tenders in that Member State]. In France, Greece and Ireland: [explanation that the Parties did not or not often meet in tenders in that  Member State].

48    For the other EEA States, the bidding data submitted by the notifying party for the years 2008- 2012 in metro projects is as follows: in Denmark, [explanation that the Parties did not or not often meet in tenders in that Member State]; in the Netherlands, [explanation that the Parties did not or not often meet in tenders in that Member State]; in Portugal, [explanation that the Parties did not or not often meet in tenders in that Member State]; in Romania, [explanation that the Parties did not  or not often meet in tenders in that Member State]; in Austria, [explanation that the Parties did not or not often meet in tenders in that Member State]; in Belgium, [explanation that the Parties did not or not often meet in tenders in that Member State]; in Bulgaria, [explanation that the Parties did not or not often meet in tenders in that Member State]; in the Czech Republic, [explanation that the Parties did not or not often meet in tenders in that Member State]; in France, [explanation that the Parties did not or not often meet in tenders in that Member State]; in Greece, [explanation that the Parties did not or not often meet in tenders in that Member State]; in Hungary, [explanation that  the Parties did not or not often meet in tenders in that Member State]; in Italy, [explanation that the Parties did not or not often meet in tenders in that Member State]; in Poland, [explanation that the Parties did not or not often meet in tenders in that Member State]; in Sweden, [explanation that the Parties did not or not often meet in tenders in that Member State]; in Norway, [explanation that the Parties did not or not often meet in tenders in that Member State]. In Ireland, Latvia, Lithuania, Slovakia, Slovenia and Estonia, [explanation that the Parties did not or not often meet in tenders in that Member State].

49    See responses to questions 35, 36 and 37 of the Questionnaire to Customers.

50    The projects concerned signalling systems in Finland for a value of EUR […] and EUR […].

51    See responses to question 45 of the Questionnaire to Customers.

52    See responses to question 43 of the Questionnaire to Competitors.

53    See responses to question 43.2 of the Questionnaire to Competitors.

54    See responses to question 13 of the Questionnaire to Customers.

55    See responses to question 49 of the Questionnaire to Customers.

56    See responses to question 48 of the Questionnaire to Customers.

57    See responses to questions 46 and 47 of the Questionnaire to Competitors.

58    See responses to question 52 of the Questionnaire to Customers.

59    See responses to question 50 of the Questionnaire to Competitors.

60    See responses to question 54 of the Questionnaire to Customers.