EC, May 3, 2005, No M.3178

THE COMMISSION OF THE EUROPEAN COMMUNITIES,

Having regard to the Treaty establishing the European Community, Having regard to the Agreement on the European Economic Area, and in particular Article 57 thereof, Having regard to Council Regulation (EC) No 139-2004 of 20 January 2004 on the control of concentrations between undertakings (1), and in particular Article 8(1) thereof, Having regard to the Commission's decision of 23 December 2004 to initiate proceedings in this case, Having regard to the opinion of the Advisory Committee on Concentrations (2), Having regard to the final report of the Hearing Officer in this case (3), Whereas:

(1) On 4 November 2004, the Commission received a notification of a proposed concentration by which the German undertakings Bertelsmann AG ("Bertelsmann"), its solely controlled subsidiary Gruner+Jahr AG & Co. KG ("G+J"), and Axel Springer AG ("Springer"), acquire joint control of the German undertaking NewCo ("NewCo") by way of purchase of shares in a newly created company constituting a joint venture. Bertelsmann (and G+J) and Springer are collectively referred to as "the Parties".

(2) On 29 November 2004 the German competition authority, the Bundeskartellamt, informed the Commission that the proposed concentration would threaten to affect significantly competition, either in the German market for rotogravure printing, or, in the alternative in the German market for time-critical print products, in particular magazines.

(3) By decision dated 23 December 2004, the Commission found that the notified operation raised serious doubts as to its compatibility with the common market and the functioning of the EEA Agreement. The Commission accordingly initiated proceedings in this case pursuant to Article 6(1)I of Regulation (EC) No 139-2004 (hereinafter "the Merger Regulation") and decided that it should, by virtue of Article 9(3)(a) of the Merger Regulation, itself deal with the aspects raised by the German competition authority.

I. THE PARTIES

(4) Bertelsmann is an international media company. Its printing activities are concentrated in its subsidiary Arvato AG ("Arvato"), which controls the German rotogravure printer maul-belser in Nuremberg, the offset printer Mohn Media in Guetersloh and various other printers in Europe, such as the rotogravure printers Eurogravure SpA in Italy and Eurohueco SA in Spain. In addition, Arvato plans to start up a new rotogravure printing facility in Liverpool (UK) in the next two years. Furthermore, Bertelsmann's solely controlled publishing arm G+J, active in the publishing, printing and distribution of newspapers and magazines, has two rotogravure printing facilities in Germany, located in Itzehoe (near Hamburg) and Dresden.

(5) Springer is active in the publishing, printing and distribution of newspapers and magazines, and holds shares in television and radio broadcasters. Springer operates two rotogravure printing facilities in Germany, namely in Ahrensburg (near Hamburg) and in Darmstadt. It also operates three off-set printing facilities which print exclusively newspapers.

II. OPERATION

(6) The notified concentration concerns the creation of NewCo, a joint venture between Bertelsmann, G+J and Springer, which will be established with headquarters in Hamburg, Germany. Following the transaction, Bertelsmann and G+J will each hold an interest of 37.45% in NewCo and Springer will hold the remaining 25.1%.

(7) The Parties will contribute to NewCo:

- their five existing rotogravure printing facilities in Germany and the planned facility in the UK, including the marketing and sales departments. Arvato's printing facilities in Spain and Italy will remain with Arvato and are not part of the notified transaction.

- the shares of maul-belser in maul + co. - Chr. Belser Studios GmbH (100%), maul + co. - Chr. Belser Klebebindung GmbH (100%), mbs Pforzheim GmbH (50%), LOG Logistik GmbH (46%) and G+J in the GWL - Gruner Druck Weiterverarbeitung und Logistik GmbH;

- the printing volume as currently agreed with third-party publishers. (8) According to a "Framework Printing Agreement" concluded between Bertelsmann, G+J, Springer and NewCo, the joint venture will print the magazines of G+J and Springer for the next [...]* years. Following the initial [...]* period, NewCo has also a matching right for the printing of [...%]* of a considerable share of the magazines of G+J and Springer, in particular the German magazines. The Framework Printing Agreement has been concluded for the period until [...]*.

(9) During the Commission's procedure Bertelsmann acquired the paper wholesaler Euro-Papier NV ("Euro-Papier") via the printer maul-belser. The concentration has been cleared by the Bundeskartellamt. Euro-Papier will be contributed to the proposed joint venture and is therefore part of the notified concentration.

III. CONCENTRATION

(10) NewCo will be jointly controlled by Bertelsmann and Springer. NewCo's board will consist of the CEO and three non-executive directors. G+J has the right to appoint the CEO and one non-executive director, and Bertelsmann and Springer will appoint one non-executive director each. Decisions relating to a number of strategic decisions, including the annual business plan, the budget and the investment planning, require the prior approval of 75% of the shareholders' meeting. Therefore Bertelsmann and Springer will each enjoy veto rights relating to the strategic commercial behaviour of NewCo.

(11) The notified concentration constitutes a full function joint venture. NewCo will be a separate legal entity with its own printing facilities, machinery, assets, personnel and customer base. Although NewCo, under the terms of the Framework Printing Agreement with its parents, will print the magazines of G+J and Springer for the next [...]* years, its full-function character will not be affected by a strong dependency on sales to its parents. The printing volumes performed to the benefit of the parent companies under the Printing Agreement do not account for more than [...%]* of NewCo's total printing capacity. Therefore NewCo will be geared to play an active role in the market and perform, on a lasting basis, all the functions of an autonomous economic entity.

(12) The Parties argue that the Framework Printing Agreement, including the right of the joint venture to print the magazines of G+J and Springer for the next [...]* years and the matching right for [...%]* of a certain magazine printing volume until [...]*, is an integral part of the concentration as it was decisive for the valuation of the joint venture and of the shares of each joint venture partner and as [...]*.

(13) Under the Merger Regulation an agreement is an integral part of the concentration if it carries out the main object of the concentration, such as an agreement relating to the sale of shares and assets of an undertaking. This applies clearly to the contribution of the five printing facilities (including personnel, etc.) which are therefore an integral part of the transaction. However, the Framework Printing Agreement is not the main object of the concentration and does not have any impact on the position of the proposed joint venture on the market. It only concerns the relationship with the parent undertakings. The subjective considerations of the Parties concerning the valuation of their shares (and the future profits of the proposed joint venture) do not change this. [...]* The Framework Printing Agreement therefore is not to be qualified as an integral part of the concentration, but the admissibility of its individual provisions has to be assessed under the Commission Notice on Restrictions directly Related and Necessary to Concentrations4. According to this Notice, supply obligations can normally be justified for a transitional period of up to five years.

IV. COMMUNITY DIMENSION

(14) The undertakings concerned have a combined aggregate world-wide turnover of more than EUR 5 billion (5). Bertelsmann and Springer each have a Community-wide turnover in excess of EUR 250 million, but they do not both achieve more than two-thirds of their aggregate Community-wide turnover within one and the same Member State. The notified operation therefore has a Community dimension.

V. RELEVANT MARKETS

A. Relevant Product Markets

(15) The notifying Parties submit that there is a uniform market for high-volume illustration printing (exceeding a volume of 200,000 copies), consisting in the printing of catalogues for mail order companies, magazines and advertisements in both rotogravure and heatset web offset (hereinafter "offset") method. In this respect, the Parties are of the opinion that offset machines with 32 pages and more per rotation are substitutable for rotogravure printing machines.

(16) In the alternative, the Parties submit that there are .sub-segments of the market. for catalogues and magazines/advertisements. The market for "catalogues" would include all printing products with a volume of more than 2 million copies, in particular mail order catalogues but also magazines and advertisements exceeding this threshold. The Parties submit that the rotogravure method is normally used for the printing of such volumes. The market for "magazines and advertisements" would include all printing products with a print run of more than 200,000 copies and less than 2 million copies. The Parties submit that these products are printed interchangeably in the rotogravure and offset methods.

(17) In a decision (6) of December 2003 pursuant to Articles 6(2), 6(1)(b) concerning the Spanish market, the Commission found a specific product market for rotogravure printing, distinct from the market for offset printing services. The Commission considered that for publications to be printed in high volumes and in high quality rotogravure printing was not substitutable by offset printing, although the latter technology has made significant developments in recent years in terms of speed of printing and quality. Customers would continue to use rotogravure printing even in the event of a significant and non-transitory price increase, as a printing machine using the rotogravure method is able to print more copies of a uniform quality in less time. The Commission further considered whether there were distinct markets according to the type of product printed, namely for magazines, catalogues and advertisements. However, in that case, the product market definition could be left open in this respect.

(18) In an opinion of 2002 the French Conseil de la Concurrence (7) described distinct markets for (1) the printing of high volumes of catalogues for mail order firms, (2) the printing of magazines, and (3) the printing of catalogues and advertisements in medium or low volumes. The Conseil found for the catalogue market that the rotogravure and the offset methods are not substitutable as the high volumes involved in this market require the use of the rotogravure technology. The Conseil further found that magazines in France are also printed by offset and not exclusively in the rotogravure method. As regards the market for medium or low volumes of catalogues and advertisements, the Conseil considered that they are mainly printed in offset.

(19) With respect to the relevant product markets, the Commission's market investigation focused on the question whether and to what extent rotogravure and offset printing are interchangeable techniques and whether the different printing applications, namely magazines, catalogues and advertisements, constitute separate product markets. Since the merger takes place in Germany and has its main effects there, the analysis refers largely to data gathered from the German market participants.

1. Rotogravure vs. offset printing

1.1 Technical comparison

(20) From a technical point of view, rotogravure and offset printing use two different printing methods. In rotogravure printing, the image area is engraved relative to the surface of the image carrier, which is a copper-plate steel cylinder that is usually also chrome plated to enhance wear resistance. The gravure cylinder rotates in an ink trough or fountain. The ink is picked up in the engraved area, and is scraped off the non-image area with a steel "doctor blade". The image is transferred directly to the web where it is pressed against the cylinder by a rubber-covered impression roll, and the product is then dried.

(21) The heatset web offset printing method is a printing system which uses plates instead of cylinders as image carriers. The image and non-image areas on the plate are chemically differentiated: the image area is oil receptive and the non-image area is water receptive. Ink rollers are used to apply onto the plate an oil-based ink which adheres only to image areas of the metal plate. The image is then transferred to a rubber blanket which subsequently transfers the image to the paper or other printing substrate. In the heatset process, which is used for high-quality print products, dryers are added to the offset printing machines in order to avoid the blurring of the colours in the further processing. This is achieved by using special heatset inks, which are essentially dried after the printing process by means of brief heating.

(22) With both methods four colours, namely black, red (magenta), blue (cyan) and yellow, are used, out of which almost any other colour can be composed. Therefore, four plates or four cylinders are necessary in order to create a coloured image. In order to print recto-verso simultaneously, printing presses are usually equipped with eight cylinders or plates. To both types of printing presses can be added finishing machines to bind the paper, using the methods of saddle-stitching or perfect binding either in the printing machine (in-line finishing) or in separate finishing machines (off-line finishing).

1.2 Capacities

(23) The capacities of the individual rotogravure presses are much higher than those of offset printing presses. According to the notifying Parties, rotogravure machines can be classified into the following machine categories which refer to the width of their cylinder: 2000 mm, 2400 mm, 2650 mm, 3000 mm and 3500 mm. It appears that modern rotogravure machines are even larger: the Parties have already installed at least one machine with a width of 4320 mm. The larger offset machines are classified according to the number of pages they can print per rotation: 32, 48, 64 and exceptionally 72 pages (the largest machine type available, with only six machines installed in the EEA).

(24) The Commission assessed the capacity of the different offset and rotogravure machines on the basis of kilotonnes (kt) per year which, according to the notifying Parties, is the most appropriate parameter for comparison. Capacity data provided by the notifying Parties is shown in the following table 1:

<emplacement tableau>

(25) It may be seen from the above table that rotogravure printing machines can reach much higher capacities than offset printing machines. In order to achieve the capacity of a 3.00-metre or a 3.50-metre rotogravure press, approximately two 64-page or 72-page offset presses, or three 48-page offset presses are needed.

(26) The notifying Parties claim that the larger offset machines with 32, 48, 64 and 72 pages per cylinder rotation are in direct competition with rotogravure printing since they can be combined in order to achieve higher volumes. For example, two 32-page offset machines could be combined in such a way that they have the same output and similar economics to a 64-page offset machine.

(27) Although such so-called "twin-webs" appear to be technically possible, the economic feasibility of such a method is very questionable. The parallel use of two 32-page machines doubles both the labour cost and the pre-printing costs (for the preparation of double or multiple sets of printing plates etc.) and is likely to increase other costs. (8) More importantly, it has to be noted that five 32-page offset machines would have to be combined in order to achieve the capacity of one single, medium-sized 3.00-metre rotogravure machine with the resulting consequences on the costs. Accordingly, the combined use of several offset presses is overall not a commercially viable option in competing with rotogravure presses.

1.3 Cost structures

(28) The costs of a printing process diverge strongly as between offset and rotogravure depending on the volume of a print order, which is determined by the number of pages per issue and the number of copies of the issue to be printed. While the costs per hour of printing are higher for rotogravure, these presses have a bigger capacity and are therefore able to process large volumes faster and more cost-efficiently than offset presses.

(29) A rotogravure printing press is a significantly higher investment for a printer than an offset printing press. This is reflected in a higher depreciation and accordingly in higher so-called "Budgeted hourly rates" (BHR), which are used by printers in order to calculate the costs (and prices) for single print-orders on the basis of the hours needed for carrying out these orders. The BHR includes costs not directly related to single print orders, mainly depreciation and personnel, calculated on an hourly basis.

(30) Similarly the pre-press stage, that is to say, the production of the printing plates (offset) or cylinders (rotogravure) is more costly for rotogravure than for offset. Not only do the costs of a cylinder itself significantly exceed those for a plate, but also the preparation of cylinders is more costly than that of plates. This mirrors the fact that the engraving of the images onto the cylinder requires more work and time than the corresponding preparation of the printing plates.

(31) Offset printing presses are more limited in the number of different pages they can print in one print-run. The surface of the cylinder (in rotogravure) or the plate (in offset) determines the number of pages which can be printed in one print-run and thereby also the time and cost of the complete printing process. The number of pages to be printed determines whether the entire issue (e.g. of a magazine) can be printed without change and interruption on the same cylinder or plate. As an example, on a 64-page offset machine only 64 A-4 pages can be printed with one plate; any further page requires the preparation and the use of an additional plate. By contrast, on a 4.32-metre rotogravure press up to 192 A-4 pages can be printed on the same cylinder without any interruption of the process, leading to a much higher output of pages per hour. Print orders for a product with a high number of pages will therefore normally be processed more costefficiently on a rotogravure printing press than on an offset printing machine, since the costs of a new print-run are avoided.

(32) Rotogravure presses normally operate at a higher rotation rate than offset presses. This means that a large number of copies can be completed in a shorter period of time on a rotogravure press than on an offset press. Owing to the higher speed, a smaller total of fixed costs is eventually attributed to the individual order. For very large volumes, therefore, the advantage of being able to print in one print-run instead of two or more and the higher speed in printing outweighs the higher BHR and pre-press costs of rotogravure.

(33) It can therefore be concluded that for large print volumes rotogravure is generally economically more efficient. This economic advantage grows with increasing numbers of pages per issue and copies.

1.4 Language/country versions and personalisation

(34) An important exemption to this rule applies when one print product is printed in different language or country versions. Some industry players do not regard these different versions as separate print-runs since not all of the plates or cylinders have to be switched, when a new language or country version is to be printed (depending on the circumstances, only the plate or cylinder for the black colour in order to change the text but not the images).

(35) Several customers confirmed that even very large numbers of copies can be printed economically on offset printing machines when many language or country versions need to be produced. This is due to the fact that the exchange of plates is cheaper than the exchange of cylinders. Moreover, in many cases submitted in the market investigation, the complete order consisted of one language or country version with a very large number of copies (400,000 - 1,000,000) and several other versions with sometimes very small numbers of copies (1,000 - 260,000). For the choice between offset and rotogravure, the cost disadvantages associated with offset for the main language version have to be balanced against the cost advantages of offset with respect to the switching of plates and the lower copy-numbers for the smaller language versions. The market investigation showed that offset is usually chosen as the more economical method if there are many language switches and only few copies for the smaller language versions. In the cases when offset was used and very large copynumbers for the main language version existed (800,000 - 1,000,000), the companies had between four and thirteen different language or country versions. Rotogravure was used when only two to four language versions were needed.

(36) Another argument which might lead to a decision for offset even for large-volume orders was mentioned with respect to personalisation, meaning the insertion of the end customer's names and addresses on the cover page or the insertion of customer-focused advertisement. Such a personalisation can technically better be processed with offset.

1.5 Paper format and quality

(37) While offset is - apart from the above mentioned exceptions - economically not efficient with large print-volumes compared to rotogravure, the same rule does not necessarily apply the other way around. Rotogravure is sometimes chosen also for small volumes. This appears to be due to the factors of format and quality.

(38) While rotogravure presses use cylinders of various diameters and therefore can be adapted to any format in both dimensions, offset presses are mostly aligned on A-4-like formats (approximately 30x21 cm +/- 2 cm). A variation is only possible either with respect to the height or the width of the format but not with respect to both dimensions at once. As a result, certain extra-high or extra-large formats cannot at all be printed with the offset technique. Formats shorter or narrower than A-4 lead either to an inefficient use of the offset machine if the printer uses a smaller paper format which only covers parts of the printing plate, or to a loss of paper. In this context it is important to bear in mind that paper accounts for approximately 50% of the total production cost. Both alternatives lead to higher costs and can make rotogravure more economical even for small print volumes.

(39) It was generally confirmed by the market investigation that rotogravure produces a higher and more stable quality than offset printing. In order to achieve comparable quality with offset, heavier paper of higher quality needs to be used in order to avoid the waving of the paper which is caused by the greater volume of moisture entailed by the offset printing process. While the quality tends to decrease in offset printing with the proceeding print-run, the engraving on the copper-plate steel cylinder which is often enhanced by chrome ensures a high stability of quality.

1.6 Analysis of orders

(40) The analysis derived from the data submitted by German print customers confirms the above mentioned considerations. The graph below incorporates the individual order data submitted by catalogue and advertisement customers. Orders containing different language versions are excluded, be they offset or rotogravure. (9) Since in some instances only annual print-runs, meaning the sum of print-runs for several issues in 2003, were indicated, the data collection contains partially very high values. For the purpose of the following graph all orders of more than 1,500,000 copies and more than 450 pages were excluded. None of the excluded entries contradicts the above mentioned delineation of the rotogravure market. This led for the year 2003 to 149 individual data sets derived from customers seeking to print catalogues and advertisements. Catalogues & Advertisements

<emplacement tableau>

(41) The analysis confirms the general pattern described above, namely that rotogravure is clearly more cost-effective for large numbers of copies (more than 400.000 - 450.000) combined with high page-numbers (higher than 64). Indeed, market participants who have not submitted detailed data per order have also confirmed this result. Offset is used with smaller volumes (owing to a smaller number of copies and/or a smaller number of pages).

(42) For magazines, the data submitted by market participants complies with the above assessment. The dividing line between rotogravure and offset is even lower than for catalogues and advertisements. An analysis of the print jobs executed by the Parties. rotogravure plants and Bertelsmann's offset printing subsidiary Mohn-Media, one of the largest offset printers in the EEA, indicates the following: both rotogravure and offset techniques are used within a relatively narrow range of 200,000 and approximately 350,000 copies. However, no magazines for European third parties are printed by Mohn Media in offset with more than 32 pages per copy and with more than 360,000 identical copies. This threshold only takes account of .regular. A-4-like formats and is even lower if the format deviates from the standard format. The market investigation largely confirmed this threshold for continental Europe, (10) where the vast majority of magazines with more than 64 pages and more than 360,000 copies are printed by rotogravure. (11)

(43) This does not rule out the possibility that in exceptional cases higher volumes may be printed in offset as some examples submitted by the Parties show (12). Exceptions may be due to specific features required by the customer (in particular personalisation, split of the magazine into different signatures each with a low number of pages, and high number of inserts) for which they consider the offset technique more appropriate, or individual preferences for specific printers.

1.7 Conclusion

(44) The market investigation has shown that for the purposes of product market definition, rotogravure printing of high volume printing orders is distinct from offset printing. Rotogravure printing is mainly used for large-volume print orders, meaning print orders with a big number of copies and pages, while the use of the offset method is largely restricted to smaller volumes. The market investigation has confirmed that offset printing does not constitute a competitive constraint for rotogravure printing of magazines with more than 64 pages and more than 360,000 copies or for catalogues and advertisements with more than 64 pages and more than 450,000 copies.

2. Magazines, catalogues and advertisements

2.1 Magazines

(45) A rotogravure printing press can print magazines, advertisements and catalogues alike. However, the market investigation showed that, at least for magazines, a separate market has to be assumed, on account of time-constraints connected to the printing of these products and to special requirements with regard to finishing and distribution.

(46) Magazines can generally be regarded as more time-critical than other printing products. Some magazines need to be printed in a very short time-span because of the topicality of their content. Apart from that, short printing times are generally also required by the publishers. advertisement customers, that is to say, the companies which place their advertisements in the magazines. Publishers reported that it is a significant competitive factor for magazines to allow advertisement customers access to the magazine up to a very late stage, in order to allow these customers to flexibly react to new events and actions taken by their competitors. This requires a very quick and optimised printing process and excludes any interruption of the process which may be possible for other print products. For catalogues and advertisements this factor of time-pressure does not exist to the same extent. Even if some catalogues and advertisements might also have short agreed printing times, this depends mostly on their own organisation of production and distribution, and not so much on external factors.

(47) Magazine printing requires specific know-how with respect to finishing and the preparation of the distribution. Finishing comprises the binding of the printed paper and the attachment of promotional add-ons and inserts. The binding can be done by either of two main methods: saddle stitching and perfect binding. Catalogues are generally finished in perfect binding while magazines are finished in both methods. The distinctive element of magazine finishing relates to the method of enclosing addons and inserts.

(48) The finishing of magazines requires specific finishing machines in order to deal with inserts and add-ons of samples. Usually the printers themselves own the specific machines and place them on their printing site or close to it in order to optimize the printing process and to better respond to the time-sensitive character of magazine printing. Unlike the finishing of catalogues which is quite frequently done by third parties to whom the individual signatures of the catalogue are shipped by the different printers involved, third-party finishing companies are used less for magazines. The requirements for dealing with inserts and add-ons differ internationally.

(49) Printers who want to supply magazine publishers not only need to have the ability to respond to the time-critical nature of the print product and to the specific finishing requirements but they also need the know-how about the distribution system used. In some countries the magazine distribution system is far more complicated than the one used for catalogues and advertisements, notably in Germany. While catalogues and advertisements are regularly sent to central distribution hubs or for postal distribution, German magazines are usually transported via the press distribution channels to newsagents. The German distribution system for magazines consists of 80 - 90 geographically dispersed wholesale traders who must be supplied individually.

(50) In addition, often a number of different versions of the same issue of a magazine have to be printed. Advertising customers may choose to place advertisements or to attach inserts or add-ons only for certain "Nielsen-areas" instead for the whole of Germany. The marketing research company Nielsen provides data for different areas in Germany which give information about the composition of target groups for advertisement. Other versions of the same magazine issue, too, may differ, such as the version for reader circles or for distribution outside Germany, which may have no samples or postcards attached.

(51) The decentralised distribution and the need to distribute different versions also influence the organisation of the printing process. The printers need to organise their printing process in order to be able to match the requirements of logistics, i.e. the early completion of those copies dedicated to the wholesale traders located very far away together with those issues which the media forwarder is going to deliver along the route. In parallel, the printers have to consider the different versions to be completed at the same time.

(52) It can be concluded that in sum the requirements of magazine printing are more stringent than those for the printing of catalogues and advertisements. They differ by country, moreover. Not all printers in Europe currently offer services which match the existing specific requirements of all magazine publishers. Their current services depend on the finishing machines they use, the know-how they have concerning different distribution systems as well as the location of their printing facilities and the distance to the publisher and to the area of distribution. Since printers located in other countries than the magazine publisher have difficulties in fulfilling the described specific requirements of these publishers, the product market definition as a result follows closely the geographic market definition discussed below.

2.2 Catalogues and advertisement

(53) While magazine printing has, for these reasons, to be regarded as a separate market, the differences between the rotogravure printing of catalogues and that of advertisements are significantly smaller. There seems to be a continuum between the rather irregular and lower-volume advertisement printing and the seasonal, very high-volume main mail-order catalogue printing. Since a dividing line is difficult to draw, the notifying Parties have suggested considering only "main catalogues" as catalogues and the rest as advertisement.

(54) It is conceivable that some advertisement print products exhibit similar features with respect to time-sensitivity as magazines. Retailers for example advertise their weekly offers with which they want to react on a short-term basis to their competitors. behaviour or wait for the latest sales figures in order to calculate their special offers. However, these potentially time-critical advertisement activities require print products of low page numbers. The products would therefore constitute a business field which is served not only by rotogravure printers but also by offset printers and is therefore not in the centre of this investigation.

(55) Moreover, advertisements are distributed in a way which does not impose any specific requirement on the printers. In this respect the distribution of advertisements rather resembles the distribution system of catalogues. The finishing of advertisements is usually simple saddle stitching without the more sophisticated techniques necessary for inserts and add-ons in the case of magazines.

(56) Consequently, it is concluded that there is a distinct product market for high-volume rotogravure printing of magazines. It can further be left open whether the printing of catalogues and advertisements in rotogravure is to be regarded as a single product market or whether separate markets for advertising printing and catalogue printing have to be considered as under none of these product market definitions competitive concerns arise.

B. Relevant geographic market

(57) The notifying Parties submit that their proposed high-volume (rotogravure and offset) illustration printing market is EU-wide. As regards their proposed alternative subsegmentation of the market, the Parties submit that the market for catalogues (with more than two million copies) is EU-wide as each of the main catalogues is printed by various printers in different countries. For the market for magazine and advertisement printing, the Parties submit that each printer can supply customers at least within a radius of 700 km. The drawing of circles of 700 km around the main European rotogravure and offset printing sites would result in a picture of largely overlapping circles. The Parties conclude that these circles would lead to a chain substitution effect so that the relevant geographic market would cover the EU.

(58) In the decision M.3322 - Polestar/Prisa/Iversiones Ibersuizas/JV (13), the Commission concluded, in line with the notification in that case, that the market for the printing of publications was national in scope. In that case, it was not necessary to decide whether magazines, catalogues and advertisements constituted one single or several separate product markets. However, the Commission's conclusion on the national geographic dimension in that case was mainly based on findings of the market investigation which were particularly pertinent for magazines, such as proximity, rapidity, costs and distribution requirements.

(59) The Commission's investigation in this case has shown that the structure of supply and demand for rotogravure printing services in Germany differs considerably from the situation in most other European countries. Owing to the large rotogravure printing capacity available in Germany which accounts for almost 50% of the total capacity installed in the EU (14), there are considerable exports of printing services, in particular to France and the UK. By contrast, German customers purchase rotogravure printing services abroad only to a rather limited extent. However, the number and volume of print jobs handled by foreign rotogravure printers for German customers varies among the different product markets described above.

(60) The merger has its main effects in Germany where it combines five rotogravure printing facilities. Therefore, the following analysis will use as a starting point the conditions of competition in the two rotogravure printing markets (for magazines and catalogues/advertisements) in Germany before assessing whether and to what extent these conditions differ from those in other geographic areas.

1. Magazines

(61) As to magazines the Commission's investigation has revealed several elements of evidence that this market is currently limited to Germany. First, only a very small part of German magazines is so far printed by foreign printers. Second, this low import ratio is due to a series of technical and economic factors in favour of domestic printers, namely the time-critical nature of many magazines, the specificities of the German distribution system and the particularities of German magazines with respect to finishing requirements.

(62) Imports account for less than 4% of the total German demand for rotogravure magazine printing, and a very large share of these imports stems from Burda's printing sites in France and Slovakia. There is only one German magazine, accounting for far less than 1% of the German market for rotogravure magazine printing, which is printed abroad by a foreign printer.

(63) One of the reasons for the almost exclusive use of German printers by German magazine publishers is the time-critical nature of these products. The risk of delays in delivery increases with the distance between the printing site and the distribution area, for example due to traffic jams or technical problems of a truck. A delayed release of a magazine not only leads to significant financial losses for the publisher but also damages the image of the magazine. The publishers are therefore very much inclined to minimize such a risk by choosing printing companies within Germany, - in the area where the magazines are to be distributed.

(64) The notifying Parties submitted a study using the magazine Der Spiegel as an example to show that even the tight deadlines for this magazine could be met when printed at sites of competitors located close to the German border in the Netherlands and France (two sites), and at a planned site in Poland. Der Spiegel is currently printed in Itzehoe near Hamburg, and in Dresden. While the final content datareaches the printer on [...]*, the magazine needs to be available on the points of sale in the whole of Germany on Monday morning, and in some cities as early as [...]*. It appears that the choice of two printing sites (both operated by G+J) was also motivated by the large printing volume and the aim to facilitate and secure punctual distribution.

(65) The study intended to show that the calculated transport time from the printing sites abroad would not exceed the time available for printing and delivery. It was based on a series of assumptions, such as the increased use of fast mini-trucks which would lead to significantly higher transport costs. However, even assuming that foreign printers were theoretically able to meet the time frame for magazines such as Der Spiegel, the other economical and technical constraints would not be altered. This is also illustrated by the mere fact that other magazine publishers who do not face such tight time constraints as Der Spiegel have so far almost exclusively chosen printing sites within Germany.

(66) Time-sensitivity does not, on the other hand, lead to a (regional) market narrower than Germany. There are only two German magazines printed at two different sites: Der Spiegel which had previously been printed for a long time at a single site, and ADAC Motorwelt, a membership magazine with a print run of 14 million copies which is distributed by mail. In both cases the main reason for the print split is the high volume to be printed within given time frames; the special situation of these two magazines would therefore not justify a market definition narrower than national.

(67) The supply of German publishers by foreign printers is further hampered by the specific conditions of distribution in Germany. The German magazine distribution system is comparatively complicated on account of its decentralised structure, with more than 80 regional distribution hubs as already described above. In other countries, such as France, the distribution is organised over a centralised distribution hub. In combination with the need to print a number of different versions, and in particular to allow advertising customers to place their advertising only in specific Nielsen-areas, the decentralised distribution requires printers to organise the printing process according to the truck routing and departure, for example by bringing forward the printing of some regional editions for remote areas and to complete at the same time those editions which have to be delivered by the media forwarder along the route. Foreign magazine printers cannot simply enter the German market for magazine printing by using their existing methods of magazine printing. It was confirmed by the market investigation that it would take a considerable time-period and close cooperation with the publishers to acquire the specific know-how and adjust the printing process accordingly.

(68) Another factor which distinguishes German magazines from magazines in other countries is the different and more complex use of add-ons, and hence the distinctive way of finishing. While in many other countries additional advertising products which are added to a magazine are usually wrapped into a transparent plastic envelope, this practice simply does not exist in Germany. The market investigation showed that German magazines are generally characterised by a particularly high number of advertisement inserts which are loosely inserted at specific pages of a magazine, and add-ons which are regularly firmly attached to the magazine. Printers abroad would therefore have to adjust their machinery to this habit in order to be able to supply German magazine publishers.

(69) On the basis of these findings it is concluded that the geographical scope of the market for the rotogravure printing of German magazines is limited to Germany.

(70) As to the other countries, such as France and UK, the market investigation has shown that imports of rotogravure magazine printing services from Germany are significantly higher than the other way around. The reason for this is the historically larger capacity located in Germany. This apparently led to some extent to differing preferences and a higher readiness of magazine publishers in the other countries to print abroad than is the case with German publishers. However, the exact definitions of the geographic market for these countries - with the exception of Germany - can be left open in this respect since even the narrowest possible delineation of the geographic markets as national markets does not raise any competition concerns in these countries.

2. Catalogues and advertisement

(71) As to catalogues, the market investigation showed that print orders are regularly split among several printers in order to ensure security of supply and a timely delivery of the required high volumes. It was broadly found that not only catalogue customers in countries other than Germany import printing services, mostly from Germany, but also that German customers regard foreign printers as viable alternatives.

(72) Since catalogues are of clearly less time-critical nature than magazines (catalogues are printed within periods of usually between 2 and 6 weeks), also printers located outside Germany are regularly used by German catalogue editors. Many German catalogue publishers place at least parts of their printing orders with foreign printers, in particular Roto Smeets (Netherlands), Quebecor (France), Mondadori and Rotocalcografica (both Italy), and Ringier (Switzerland) as well as subsidiaries of German printing companies located abroad, such as Burda in France and Slovakia, and Rotoalba in Italy (a subsidiary of TSB). According to the market investigation and the data delivered by the printers, approximately [25-30%] of the German catalogue volume is printed abroad, including [15-20%] in Burda's French and Slovakian printing facilities.

(73) The majority of German catalogue print customers who submitted data within the market investigation currently use printing companies abroad or have already done so in the past. The majority of these customers, moreover, indicated that they could increase their share of printing abroad. The shares of the individual printing volumes that these customers have placed with printing companies abroad range between 10% and 51%. Approximately one-third of the customers giving a reply indicated that they could not print abroad or would only use subsidiaries of German printers located abroad (France, Slovakia and Italy). The latter point in particular shows that with respect to catalogues the sheer distance between printer and distribution area is not as important for the choice of a printer as is the case in magazine printing. This is in line with the generally lesser time-sensitivity of these printing products.

(74) One important issue in the choice of the printer is apparently the language. During the preparation of the printing process many technical details have to be co-ordinated and precisely adjusted between the printer and the customer. In order to avoid any risk arising from difficulties in communication, German printing customers obviously prefer to have technical assistance provided by the printer in German language.

(75) This language question might be the reason why German customers to some extent prefer German subsidiaries abroad over foreign printers. However, the investigation confirmed that the major printers in the neighbouring countries (and Italy) who have successfully acquired German catalogue publishers, provide technical assistance in the German language (at least Roto Smeets, Quebecor and Mondadori; Ringier is located in the German-speaking area of Switzerland), and most of them also have a sales office in Germany. This linguistic presence matters, as customers usually come to the printing facility for the acceptance procedure, often accompanied by a member of the German sales office.

(76) As a result it can be concluded from the responses in the market investigation and from the ordering behaviour of German catalogue customers that printers in countries adjacent to Germany, and in Slovakia and Italy, constitute viable alternatives for German catalogue producers. Since customers from these countries also use printing services from other countries within this area - mainly Germany - the geographic market can therefore be defined as Germany plus the neighbouring countries (France, Belgium, Netherlands, Luxemburg, Switzerland, Austria, Czech Republic, Poland, Denmark) together with Italy and Slovakia, covering the large printers in these areas such as - beside the German printers - Quebecor, RotoSmeets, Mondadori, Ilte, Rotocalcografica and Ringier. The UK is not part of this market since German catalogue publishers unanimously did not consider the printers located in the UK to be a viable alternative.

(77) This market definition is compatible with Decision M.3322- Polestar/Prisa/Iversiones Ibersuizas/JV, where the Commission has concluded that there is a national Spanish market for rotogravure printing of publications, including catalogues and advertisements. It appears that the geographic situation in Spain differs significantly from that part of Europe which is under consideration in this case, since there are no rotogravure printers in the adjacent areas, i.e. Portugal and the South-West of France. For Spanish customers, the possibilities of importing rotogravure printing services are therefore very limited.

(78) Advertisement printing for German customers is apparently to a large extent carried out in Germany: imports account for around 5% of the market. This does not, however, mean that a national market can be assumed. It appears from the market investigation that the reason for the focus of German advertisement customers on domestic printers is rather the availability of sufficient capacities and suppliers in Germany. So far, German customers of rotogravure advertisement printing therefore have had no reason to turn to foreign suppliers, in particular as the motive of risk diversification applicable to catalogues, whereby very large volumes are split, does not prevail to the same extent.

(79) However, in spite of a lower import ratio for advertisement printing than for catalogue printing, German customers can easily turn to credible foreign printers having the capability and the equipment to supply such services to Germany. The printing of advertisements does not create any specific difficulties comparable to those in the magazine printing market, such as the special finishing or specific conditions of distribution. Most advertisements are finished by simple saddle stitching without any specific requirements for add-ons or other additional features. Distribution apparently does not exhibit specific difficulties and is mostly organised by the customer. Moreover, as was described above, advertisements are generally not as time-critical as magazines. Consequently, every printer abroad who already prints catalogues for German customers can be assumed to be also capable of printing advertisements to be distributed in Germany. Since most publishers of main catalogues also issue advertisements, it would in addition be easy for them to use the existing links to foreign printers for advertisement orders too. For this reason, the geographic scope of the market for rotogravure advertisement printing can be considered to be the same as for catalogues, comprising Germany, its neighbouring countries, and Italy and Slovakia.

VI. COMPATIBILITY WITH THE COMMON MARKET AND THE EEA AGREEMENT

A. Markets for Rotogravure Printing

1. Overview

(80) The proposed joint venture brings together the current leader among the German and European rotogravure printers, Bertelsmann (including the subsidiary G+J), and the third-strongest German rotogravure printer (and sixth-strongest in Europe), Springer, assessed on the basis of installed capacity for rotogravure printing in tonnes. The concentration will therefore strengthen Bertelsmann's leading position in Germany and Europe in the field of rotogravure printing. Bertelsmann's rotogravure activities in Italy and Spain will remain outside the joint venture. Those activities will be taken into account when assessing the effects of the proposed joint venture in those markets where both NewCo and the printing facilities remaining with Bertelsmann are active.

(81) The joint venture partners' capacity, production and third-party sales on the basis of the figures for 2003 are as follows:

<emplacement tableau>

(82) Outside the proposed joint venture, Bertelsmann operates two further rotogravure facilities in northern Italy: Eurogravure in Bergamo and Milan. It also runs one printing facility in Spain, Eurohueco. The facilities in Spain and Italy have a combined capacity of [200-250 kt]*.

(83) The main other players in the sector of rotogravure printing in Europe are the following:

- Schlott and TSB in Germany. These undertakings are not vertically integrated into publishing, but only operate rotogravure printing facilities (as well as heatset web offset). Schlott is the largest printer in Germany after Bertelsmann (and the third-largest in Europe) and operates four rotogravure sites in Germany, located in Hamburg, Freudenstadt, Nuremberg and Landau, with a total capacity of [500- 550] kt. TSB operates two rotogravure printing facilities in Germany, located in Moenchengladbach and Munich with a total capacity of [200-250] kt as well as a smaller rotogravure printing facility in Italy, called "Rotoalba" and located close to Turin, with a capacity of [50-75] kt.

- The publishers and printers Burda and Bauer in Germany. Whereas Burda offers a significant portion of its printing capacity to third parties, Bauer uses its printing facilities nearly exclusively for the captive printing of its magazines. Burda operates a large printing facility in Offenburg in Germany with a capacity of [150-200] kt, a further rotogravure printing facility in Vieux-Thann (France) with a capacity of [100-150] kt and a small rotogravure printing facility in Bratislava (Slovakia) with a capacity of [25-50kt]. Bauer operates a large rotogravure facility in Cologne (Germany) with a capacity of [150-200] kt, and two sites in Poland, one in Ciechanow with a capacity of 101 kt, according to the estimates of the Parties, and a planned facility in Wykroty.

- The non-vertically integrated printer Quebecor with five rotogravure printing facilities in France (located in Blois, Corbeil, Lille, Mary-sur-Marne and Strasbourg) with a total capacity of 383 kt according to the estimates of the Parties and one facility in Belgium and Finland respectively (each with a capacity of 77 kt according to the estimates of the Parties). Quebecor is the second-largest printer in Europe, after Bertelsmann (and the joint venture).

- The non-vertically integrated printer Lenglet with one rotogravure site in France, operating only since 2002 and with a capacity of [50-100] kt.

- The non-vertically integrated printers Roto Smeets and Biegelaar in the Netherlands. Roto Smeets operates two sites in Etten-Leur and in Deventer, with a combined capacity of [250-300] kt. Biegelaar operates one small printing site with a capacity of [50-75] kt.

- The publisher and printer Mondadori which operates two rotogravure printing facilities in northern Italy, located in Melzo and Verona, with a total capacity of [150-200]* kt, which is used to a considerable extent for its captive needs. The printer Rotosud uses its capacity exclusively for the captive needs of the publishing parent company.

- The non-vertically integrated Italian printers Ilte (with a printing site in Turin and a capacity according to the estimations of the Parties of 146 kt) and Rotocalcografica, operating a rotogravure printing site close to Milan with a total capacity of [50-75] kt.

- The non-vertically integrated printer Polestar with three existing rotogravure sites in the UK, located in Scarborough, Bristol and Pershore, with a total capacity of 290 kt according to the estimates of the Parties and a printing facility under construction in Sheffield which is due to start production in 2005. Polestar further operates two printing facilities in Spain with a capacity of 159 kt according to the estimates of the Parties.

(84) The position of the Parties in the field of rotogravure is further shown by their share of the capacity for rotogravure printing installed in Germany.

<emplacement tableau>

(85) The table above clearly shows the importance of the proposed joint venture in the field of rotogravure printing in Germany. It will have more than double the total capacity of its nearest competitor and will have still more than one-and-a-half times the capacity available for third parties of its next competitor.

(86) The proposed joint venture is also the leading player in Europe. Whereas the installed annual capacity in Germany amounts to 2322 kt (19), the installed annual capacity was only around 656 kt in France and 333 kt in the UK (20) according to calculations of the Parties. The importance of the rotogravure capacity installed in Germany is further shown by the fact that the German share of the European rotogravure capacity is more than 45%. The capacity of the proposed joint venture therefore exceeds the total capacity installed in France and the UK.

(87) Players from other Member States are considerably smaller than the proposed joint venture. On a European scale, Arvato's rotogravure capacity in Italy and Spain has also to be taken into account. According to the Commission's calculations, the capacity of the proposed joint venture and of Arvato's Italian and Spanish facilities will be more than double the capacity of Quebecor, taking Quebecor's rotogravure printing facilities in France, Belgium and Finland together. The third player on a European scale is the German company Schlott with the capacity described above.

2. Structure of the relevant markets

(88) The proposed joint venture will be active in the markets for the rotogravure printing of magazines, catalogues and advertising in a number of countries within the EEA, but the most serious effects of the proposed concentration - owing to the location of the five existing printing facilities brought into the joint venture - will be felt in Germany. The Commission analysed the impact of the transaction in the different markets first on the basis of market shares. In the following paragraphs, all markets in which the proposed joint venture would lead to an addition of market shares and to a combined market share exceeding 15% will be discussed. The geographic market definition for the markets beyond Germany has been left open. For those markets, the analysis will be carried out on the basis of the narrowest geographic market definition conceivable. If no competition concerns arise on the basis of this market definition, competition concerns can be generally considered not to arise.

(89) The Parties have estimated market shares on the basis of the installed capacities of rotogravure machines, assuming an average usage of the machines of 85%, deducting the estimated intra-group sales of the vertically integrated printers, and identifying the proportion of imports and exports on the basis of the Parties. knowledge and the Eurostat statistics of the Commission. The Parties further assumed in their estimates that the competitors have a similar split between the three print products: magazines, catalogues and advertising. For their own merchant sales in the year 2003, the Parties submitted the following figures for those different print products, including only those parts of the Parties which are brought into the proposed joint venture.

<emplacement tableau>

2.1 German market for magazines

(90) On the basis of the results of the market investigations, the Commission calculated the market shares for the German merchant market for rotogravure printing of magazines in line with the general approach of the Parties, namely on the basis of tonnes of paper used for the printing of magazines for third parties. According to these calculations, the joint venture will be the clear market leader, with a market share of around [45-50%]* in the German merchant market for rotogravure printing, i.e. excluding printing for in house publishers. The next players are TSB and Schlott with around [20-25%]* each and Burda with a share of [0-5%]. Imports account for [0-5%], which are to a very considerable extent supplied by Burda's printing facilities in Vieux-Thann (France) and Bratislava (Slovakia). Apart from those imports, only one German magazine is printed abroad, by the Dutch printer RotoSmeets.

(91) On the basis of this market structure, competition concerns can be ruled out from the outset, and therefore the market for the rotogravure printing of German magazines will be discussed in detail below.

2.2 Markets for magazines in other EEA Member States

(92) The merger does not lead to a situation which could give rise to competition concerns in other Member States of the EEA in the markets for rotogravure printing of magazines. In such markets, defined on a national basis as the narrowest market definition conceivable, market shares exceed 15% only in the UK, Austria and the Czech Republic, on the basis of the estimates submitted by the Parties for 2003.

(93) According to the information submitted by the Parties, the proposed joint venture reaches a combined market share of [20-30%]* (equalling [40-50 kt]*) in the UK market for rotogravure printing of magazines (having a total market volume of 178kt in 2003), with an additional 1% market share achieved by Bertelsmann's remaining activities. However, the overlap is de minimis, as Springer's market share only accounts for [0-1%]* (equal to [0-1 kt]*). Polestar is much stronger than the proposed joint venture, having a market share of [40-50%]* (equal to [50-100 kt]*). Bertelsmann is currently in the process of constructing a new rotogravure printing facility in Liverpool which will start operations in 2006/2007. The capacity will be [150-200 kt]*, of which [100-150 kt]* are reserved for a large printing order for magazines for a UK customer. Although the printing facility is planned to be brought into the proposed joint venture, a possibly increased market presence in future due to the new plant cannot be attributed to the creation of the joint venture. The construction of the plant was planned to be undertaken by Bertelsmann alone, so that the proposed concentration is not causal for any potential increase in market share due to the construction of the new plant in Liverpool. Even considering a purely speculative increase in market share in the future, the considerations mentioned above are valid in showing that the increment contributed by Springer's presence on this market is de minimis; Polestar is currently the market leader by a wide margin, and will remain a very strong competitor, account being taken of the fact that Polestar is currently constructing a new rotogravure printing facility with a large capacity in Sheffield which will be operational in 2005.

(94) In Austria, the proposed joint venture would have a market share of [60-65%]*, equivalent only to a volume of [10-15 kt]* (Bertelsmann's remaining activities have an additional market share of [1-5%]*). However, Springer is not active there, so that no overlap in market share exists. In the Czech Republic, Bertelsmann has a market share of [25-30%]* in magazine printing (equal to a volume of [5-10 kt]*), but Springer does not add any further market share and the competitor, Nase Vojsko, reaches a share of [30-35%]*.

2.3 Markets for catalogues and advertisements

(95) In the market for rotogravure printing of catalogues and advertisements, the market shares of NewCo would amount to [20-25%]* on a European market including Germany, the neighbouring countries and Italy and Slovakia according to the estimates of the Parties on the basis of the volumes for 2003. On the basis of distinct markets for catalogues and advertisements, the proposed joint venture would have a share of [15- 20%]* in a market for catalogue printing and of [20-25%]* in advertisement printing. For Bertelsmann's rotogravure printing activities being kept outside the joint venture, an additional [1-5%]* has to be added to each of these market shares. On separate markets as well as on a combined catalogue and advertisement market, Schlott and Quebecor would follow close behind with market shares of between 13% and 14% and TSB with approximately 11%. On such a European market, competition concerns therefore arise neither for a market combining catalogue and advertising prints nor for distinct markets for catalogue and advertisement printing21.

(96) The merger does not raise competition concerns for the other markets within the EEA for catalogue and/or advertising printing, even if those markets are delineated as national in scope using the narrowest conceivable market definition. According to the information submitted by the Parties for the UK, no critical market shares are reached on the markets for rotogravure printing of catalogues and/or advertisements. The proposed joint venture would have a share of [20-25%]* in a market for the printing of catalogues and advertising, and of [30-35%]* on a market for catalogues and [15- 20%]* for advertisement printing. Polestar is the leader in these markets with market shares between 42% and 46%.

(97) If the market for Sweden is defined as national in scope using the narrowest conceivable geographic scope, the proposed joint venture would have a market share of [25-30%]* on a market for catalogue and advertising printing and, assuming distinct markets, [35-40%]* on a market for catalogues and [20-25%]* for advertisement printing. Quebecor follows closely with a share of the combined market of 28% and with shares of 25% for a catalogue market and 32% for an advertisements market, where it is the leader. Even if the market share for a distinct market for catalogue printing could be considered high, it has to be noted that Springer only accounts for an increment of less than [0-5%]*, equal to a volume of less than [... kt]* in a market which has a volume of only 17 kt overall. Given the small total market volume, even one printing order may significantly change the picture without conferring decisive market power on the market participants. If Norway and Finland are included in the market, market shares for such a Nordic market for catalogue and advertising printing would drop considerably, to far less than 20% on a combined market and a market for advertisement printing, and considerably less than 30% on a market for catalogue printing.

(98) It can therefore be concluded that no competition concerns arise from the proposed concentration in the markets for catalogue and advertising printing, howsoever the geographic scope of these markets is delineated.

3. Potential competitive harm to customers in the German market for rotogravure printing of magazines

(99) As was already mentioned above, the proposed joint venture will be the clear market leader in the German market for the printing of magazines. The shares and tonnes of the merchant market for the rotogravure printing of third-party magazines for the year 2003 are as follows:

<emplacement tableau>

(100) As already indicated above, the Parties' main competitors are TSB and Schlott, with a market share of [20-25] % each, and Burda (Germany only) with a market share of [0-5] %. If imports from Burda's printing sites in Vieux-Thann and Bratislava are taken into account, Burda's market share increases to [5-10] %. Bauer cannot be considered a competitor in magazine printing for third parties. Bauer uses its capacity nearly exclusively for its captive needs and does not print a single third-party magazine. The Parties claim that Bauer made an offer for the printing of a third-party magazine some years ago. However, the Commission does not have any indication that Bauer is currently considering the printing of third-party magazines.

(101) The market investigation did not allow an assessment of the market shares over several recent years. In general, it may be assumed that market shares are relatively stable. Although some switches of printers occurred in recent years, in general the relationship between magazine customers and rotogravure printers is quite stable, sometimes leading to the result that a printer prints a magazine for several decades. However, the Parties submitted that the relationship between the merchant market and the captive printing has changed. In 2004, Springer acquired the magazine TV Digital for the German broadcaster Premiere whose printing volume of [1-5 kt]* has therefore become captive. Furthermore, G+J has very recently acquired the publisher Motorpresse (22) whose rotogravure printing volume for the magazines [...]* and [...]*, of close to [... kt]*, will become captive. Whereas these acquisitions by the Parties at the same time reduce the volume of the merchant market as well as their market share, the recent acquisition of the publisher Milchstraße by Burda only reduces the volume of the merchant market, but might even increase the Parties' market shares. Owing to the lack of figures for 2004-5, the exact impact of these acquisitions on the market shares cannot be calculated, but it can be concluded that the overall volume of the merchant market and the supply of the Parties have been reduced.

(102) Furthermore, the Parties waived the matching right [one of the Parties had with regard to an important magazine printing volume. The waiver of this matching right will make this printing volume contestable by third-party printers after the end of the contract and will thereby make sure that this part of the Parties. position, too, is under a competitive threat from third-party printers and that this is not an unchangeable part of the Parties' market share.]*

(103) In the light of the high market shares and despite the mitigating factors, serious competition concerns could be raised for the German market for rotogravure printing of magazines. In such a market, customers - the publishers of magazines - could be harmed if the joint venture were able to profitably raise prices and if customers were not able to counter such price increases by switching to other suppliers on account of a lack of available capacity or an insufficient number of suitable competitors.

3.1 Available alternative capacity of competitors

(104) Customers could counter a price increase by the proposed joint venture if they were able to replace a significant volume of the quantities purchased from the Parties with supply from competitors so that a price increase would be rendered unprofitable for the Parties. As is shown in table 4 above, the volume supplied by the Parties to the merchant market amounted to [150-200 kt]* ([100-150 kt]* for Bertelsmann; [45-50 kt]* for Springer) in 2003. It may further be considered that a volume of around [10-15 kt]* of this supply has become captive in the meantime owing to the acquisition of publishing houses by the Parties so that no customer can be harmed by this volume. Taking this into account, the volume which the Parties supplied to the merchant market totalled [100-150 kt]*.

(105) The market investigation did not give any indication that the demand for the printing of magazines would increase in future. After a boom around the year 2000, demand for magazine printing decreased, largely on account of a reduction in advertising pages, a decrease in the number of copies per issue and the increasing importance of the Internet for the exchange of information and for advertising. The Commission therefore examined the question whether competitors would be able to replace a significant part of the capacity on the basis of the figures for the year 2003.

(106) In the following, the questions examined will be (1) whether competitors currently have sufficient spare capacity to replace these sales to a significant extent, (2) whether competitors could make available such capacity by shifting their capacity to the printing of magazines, (3) whether planned capacity extensions will make available further capacity and (4) whether potential competitors could contribute to making available further capacity for the printing of magazines in the event of a price increase. However, as a basis for this examination the general capacity allocation by rotogravure printers will be analysed.

Current capacity allocation by competitors

(107) The Commission analysed the way in which rotogravure printers allocate their capacities to the different print products - magazines, advertising, catalogues - in order to determine how much of their current spare capacity the competitors can use for the printing of magazines and how much of the capacity currently used for the printing of catalogues and advertising could be shifted to the printing of magazines.

(108) One of the main characteristics of the rotogravure printing industry is that it is driven by capacity. The costs of the printing press account for a very considerable share of the total costs for operating the printing facility. According to the calculations of the Parties for printing presses with a width of 3.60 - 4.32 m, the two main cost factors are personnel costs and depreciation, whereby personnel costs account for approx. 29 - 39% and depreciation of the printing press for approx. 26 - 30% of the total indirect costs (expressed in the .Budgeted Hourly Rate. as described above). These exclude costs directly related to an order such as paper, ink and transport. These direct costs are fully variable and only transitory items for the printer. The situation as to the pre-press costs which are related to the engraving of the cylinders and their further preparation for the printing process is similar to the cost calculation for the printing presses themselves, that is to say that the main factors are costs for cylinders, engraving machines and personnel. The main factor which drives the business of a rotogravure printer is therefore the incentive to use the installed capacity as fully as possible in order to recoup the fixed costs (or more precisely the indirect costs) of the printing press. In order to fill the capacity printers will normally try to achieve a mix of different print products - advertising, catalogues and magazines, as these print products have different characteristics.

(109) On the basis of the market investigation, the Commission has come to the conclusion that no 100% supply-side substitutability between the different print products can be assumed. In other words: it cannot be assumed that printers will switch their capacity from printing catalogues and advertisements entirely to the printing of magazines, or that they will use their entire annual spare capacity only for the printing of magazines.

(110) Limitations on the use of capacity for the printing of magazines arise first from the differences in periodicity, printing time and volume of the different print products. Magazines are printed periodically (weekly, fortnightly or monthly) with fixed and relatively short agreed printing times (one to ten days, at the very most). Owing to their long-term and periodical publication, they constitute the "base load" for the printing facility which fills the presses over the entire year. Differences in the printing volume for magazines only arise over the year as in certain periods - e.g. in summer - editions may have a lower number of pages due to fewer advertisements or may have a lower print-run. The longer-term and regular character of the printing of magazines is also reflected in the relatively long duration of the print contracts for magazines which normally ranges between two and five years.

(111) By contrast, catalogues for mail-order companies or tour operators, et cetera, are usually released only twice per year with very high printing volumes (both in terms of the number of copies and of pages) and longer printing times (up to several weeks). They are normally printed in May and June and from October till December and constitute a "peak load" for the printing presses. The duration of contracts for the printing of catalogues is shorter than that for the printing of magazines: it ranges normally between six months and three years (exceptionally, the Parties entered into longer-term contracts under specific circumstances, [...]*).

(112) The third category of print products, advertising, is in essence used to fill the printing capacity between the catalogue printing seasons and on the week days when fewer magazines are printed. As publishers prefer to release their magazines on certain week days, the printing of magazines varies over the week and there are days on which a comparatively lower number of magazines is printed. Print orders for advertising are normally placed on a short-term basis; the duration of contracts for advertising printing is normally between three months and one year.

(113) On account of these time characteristics, the majority of printing companies indicated that an unlimited use of spare capacity for magazine printing or an unlimited switch from catalogues/advertisements to magazines would not be feasible. Whereas magazines are printed throughout the year on certain week days in a given frequency, spare capacity may only be available in times of low demand; therefore, at times when no catalogues are printed or on days of the week when demand for magazine printing is low. Periods of low demand in the year are in particular February and March as well as August and September. Advertisement would at least be necessary to fill the smaller gaps in capacity utilisation within the week, since weekly magazines are not released on every day of the week to an equal extent. Moreover, owing to the existence of fortnightly and monthly magazines, apparently also bigger gaps have to be filled within the month by either advertisements or catalogues. In addition, the risk borne by a printer would significantly increase if it focused exclusively on the printing of magazines. In this case, breakdowns of machines could have the result that a magazine - given the rather tight deadlines for magazines in general - was not printed in time whereas the risk can be balanced if the printer also prints other print products with longer deadlines. In this case, the printer can switch the magazine printing to another press and can catch up on the printing of the advertisement or catalogue later. This flexibility in capacity allocation is considerably reduced if a printer focuses exclusively on magazine printing.

(114) Furthermore, a printer may not wish to abandon the printing of catalogues (and advertisements for filling the gap). Whereas such a printer may readily print magazines in times of low demand, a conflict between the obligation to print magazines throughout the year and the printing of catalogues may arise. If in such peak times neither spare capacity were available nor capacity were used for the printing of advertising, the printer would have to abandon the printing of catalogues in order to be able to print magazines.

(115) Owing to these constraints on the printing of magazines, the majority of the printers which replied during the market investigation stated that they would not be able to use their entire (annual) capacity for the printing of magazines, but that they needed to retain some flexibility. Three rotogravure printers indicated a maximum of 70% and one a maximum of 85% of magazine printing in their printing product mix. Only one printer indicated that its capacities were 100% variable for the printing of the different products (23). The Parties - although arguing in general for a 100% supply-side substitutability - admitted that a capacity utilisation exclusively based on the printing of magazines would not be an "optimal mix" since it would not allow the printer to balance the risk and would make production inflexible.

(116) Secondly, limitations may further arise due to limitations in the capacity of finishing machines. As was explained above, the finishing of magazines - either in saddle stitching or in perfect binding - is usually done by the printer in Germany and requires specific finishing machines adapted to the finishing of magazines. The capacity for the printing of magazines is therefore generally limited by the available capacity of finishing machines. However, the investment necessary for an extension of the finishing capacity is much less than that necessary to extend rotogravure printing capacity. Whereas the investment for a new finishing machine amounts to between EUR 2.5 and 4.5 million (depending on its capacity and whether it is a less expensive saddle-stitching machine or a more costly perfect binding machine), the price for a new, state-of-the-art rotogravure press (as a system) totals between EUR 20 and 30 million. The market investigation showed that customers may prompt the investment of printers in finishing machines whereas this is very rarely the case for an investment in rotogravure printing capacity. However, the market investigation also demonstrated that finishing machines are only acquired if a continuous utilisation of these machines can be expected, but not if they are not only utilised for one specific order or on one day per week. This may be especially relevant for those printers having a relatively small share of magazine printing in their product mix. The market investigation further showed that lack of finishing capacity in peak times can be overcome by outsourcing the finishing to independent finishing undertakings. This is in particular done for perfect binding of magazines, less for saddle-stitching of magazines. This solution is in particular chosen for peak times as the outsourcing usually entails additional costs and time for the transport of the magazines.

(117) Third, limitations on the printing of magazines may arise from the focus of the printer, whether the specific printer gives priority to the printing of catalogues and advertising or whether he gives priority to the printing of magazines. Despite the possibilities of overcoming lacking finishing capacity by investment in new finishing machines and by turning to independent finishing undertakings in peak times, it would not be realistic to assume that those rotogravure printers which have their current focus on the printing of advertising and catalogues would make their capacity available up to the general limit of 70 - 85% for magazine printing and thereby also largely abandon their traditional business.

Capacity of German competitors

Currently available spare capacity of competitors

(118) On the basis of the foregoing considerations, the Commission analysed how much spare capacity can be made available by competitors for the printing of magazines. Taking into account in particular the seasonality of the different print products, the Commission assumed that a careful approach should include spare capacity of the competitors for the printing of magazines only up to an annual capacity utilisation of 95%. Above that level, it cannot be safely assumed that spare capacity could be used throughout the year for the printing of magazines.

(119) Capacity utilisation has been quite high in this industry in recent years. The German competitors Schlott, TSB and Burda indicated a capacity utilisation of 90% or higher for 2003 or 2004. As regards the available capacity of competitors, the Commission included Burda's printing facility in Vieux-Thann as part of Burda's overall capacity, as Vieux-Thann is located at only 35 km from the German border and at 130 km from Burda's printing headquarters in Offenburg. Both sites therefore are regarded as a single production entity.

(120) On the basis of a maximum capacity utilisation of 95% and the figures submitted for 2003, it appears safe to assume a spare capacity for magazine printing of the German competitors of 17 kt.

Supply-Side Substitutability

(121) Magazine customers could also turn to competitors of the joint venture if German rotogravure printers - irrespective of their current spare capacity - were to make additional capacity available for the printing of magazines instead of printing advertising or catalogues. The shift of capacity to the printing of magazines does not automatically result in a reduced capacity for advertising and catalogue printing and thus lead to a price rise in these markets. Whereas the scope of the market for magazine printing is limited to Germany, the geographic market for catalogue and advertising printing is defined as Germany, its neighbouring countries, and Italy and Slovakia. German advertising and catalogue customers are therefore not limited to German rotogravure printers, but can turn to suppliers abroad such as RotoSmeets, Quebecor, Ringier, Mondadori and Ilte. This possibility for catalogue and advertising customers to turn to printers outside Germany is also confirmed by the lower market shares of the proposed joint venture in these markets as indicated above. (24)

Possibility for switching to magazines

(122) The Commission assessed the possibility of German rotogravure printers to switch their capacity currently used for the printing of catalogues and advertising on the basis of the considerations outlined above. The Commission therefore relied first of all on the figures provided by the competing printers themselves; thereby taking account of their individual limitations due to the lack of finishing capacities and their current printing focus. The competitors which replied by stating a figure indicated switching rates of up to [15-20%] of their total capacity. One German printer did not provide any figure; as a cautious approach and in line with the other results of the market investigation, the Commission assumed that this printer, who currently has a comparatively low share of magazine printing, could dedicate another [10-15%] of its capacity to magazine printing. The Commission therefore did not assume that the competitors could in general switch up to 70% of their capacity to the printing of magazines which were indicated as the general maximum for the printing of magazines in the market investigation. In a cautious approach, the Commission remained far below this figure for those printers which indicated individual constraints due to the lack of finishing machines or a focus on the printing of catalogues and advertising.

(123) On the basis of such estimations, the entire annual capacity which German printers could make available for magazine printing would be around 130 kt. This additional capacity of 130 kt which the other three German rotogravure printers could make available by switching away from catalogue and advertising printing would account for a very large share of the entire capacity used by the Parties for printing third parties' magazines ([150-200 kt]* in 2003).

(124) Furthermore, it has to be underlined that the estimate for the possibility of switching seems to be rather conservative, as it takes account of the current finishing capacity and the current focus of the printers. As was explained above, the market investigation confirmed that printers may acquire additional finishing capacity if this is needed for a major new order, or may use third-party finishing capacity in the interim and may also shift their focus at least in the medium term. Taking account of these factors, it would appear to be possible for those printers to increase the maximum capacity for magazine printing beyond the share actually assumed, at least in a medium-term perspective, if this were required by major new orders for magazine printing. Such a perspective seems a reasonable assumption, as the duration of contracts for magazine printing usually ranges from two to five years so that such contracts will only come on the market one by one over a period of several years.

Incentives for a switch to magazine printing

(125) It would not be sufficient that competitors had the possibility to switch their capacity to magazine printing (and thereby defeat the attempt of a price increase by the joint venture for magazine printing) unless they also had the incentives for doing so.

(126) In order to analyse the incentives of printers to switch their capacities to the printing of magazines, the Commission calculated contribution margins of the Parties for the different print products on the basis of all their print orders in the year 2003. The margins varied very significantly within the categories, and they even varied within the same order for the printing of different issues. This results in particular from the fact that the margins are based on actual costs so that higher costs for the printing of a specific issue due to technical problems, a larger number of inserts, a suboptimal exploitation of the size of the printing machine, etc. may lead to a reduced contribution margin although the printing of all the issues of a magazine is based on one price and one order.

(127) The Commission therefore calculated weighted averages for the contribution margins for the different print products for the three printing companies involved for all their print orders in the year 2003. These are as set out in the following table:

<emplacement tableau>

(128) As can be seen from the table above, printing of advertisements gives by far the lowest contribution margin whereas catalogues have the highest margin ([...]*). It would therefore be economically rational to assume that printers would switch their capacity away from the printing of advertisements and towards the printing of magazines on account of the higher profitability of magazine printing.

(129) However, the argument may not be as straightforward for the peak periods of catalogue printing. In these periods, as outlined above, only a little advertisement is printed but the main part of the capacity is taken up by the printing of catalogues. Nevertheless, magazines - as products periodically printed - also have to be printed in these periods. It can therefore only be assumed that a printer will switch his capacity to the printing of magazines above the volume used for advertisement printing in those periods too, if the margins for the mix of catalogue and advertisement printing (catalogues in the peak times, advertisement to fill the gaps) is less profitable than the printing of magazines throughout the year. As shown in the above table, the weighted averages for the contribution margin of catalogue and advertisement printing combined is lower than the contribution margin for magazine printing. [...]*.

(130) It can therefore be concluded that it would be economically rational for a printer to switch (1) from advertisement to magazine printing in any case and (2) also from a mix of advertisement and catalogue printing (catalogues in the peak times, advertising to fill the gaps) to magazine printing.

(131) This conclusion, reached on the basis of the Parties' data, is confirmed by the market investigation. For the other printers, the highest margins were either reached with magazine printing or the margins for magazine printing were close to the catalogue margins whereas advertisement printing was much less profitable. It can therefore be assumed that magazine printing is more profitable than a mix of catalogue and advertisement printing.

(132) Magazine printing has a further advantage which may constitute an incentive for printers for a switch to magazine printing. As magazines are printed periodically on the basis of longer-term contracts, they constitute a base load for the printing capacity. If the base load is increased, the printer has to fill less capacity with short-term orders, in particular advertisements. As a consequence, a higher share of magazines in the product-mix significantly reduces the printer's risk of not fully utilizing the existing machines and the effort of acquiring additional orders to fill the gaps.

(133) It can therefore be concluded that the German competitors Schlott, TSB and Burda would have the possibility of switching capacity from advertisement (and catalogue) printing to magazine printing and would also have the incentives to do so. The volume which could be made available would cover a very large share of the volume currently used by the Parties for the printing of magazines.

Additional capacity due to capacity increases of German competitors

(134) Furthermore, the profitability of a potential price increase for magazine printing by the Parties is likely to be thwarted by their competitors' ongoing or planned installation of new and more powerful presses which result in a net increase of rotogravure capacity available for German customers.

(135) All three German competitors, TSB, Schlott and Burda are currently in the process of installing new rotogravure presses or are planning to do so by the end of 2007. To some extent the new presses replace older presses which are to be dismantled subsequently. Nevertheless, the installation of new presses is likely to lead to an increase of net capacity as the new presses usually have a higher capacity than those old presses. And even if a printer intends to gradually dismantle the same volume of capacity as will be installed, so that there is no net increase in capacity, the printer concerned will still be able to delay the dismantling of the old presses if it expects a rising demand.

(136) On this basis, the Parties. three main competitors in Germany, Schlott, TSB and Burda, are planning to increase their net capacity by at least 50 kt over the next two to three years. This figure does not take into account their additional possibility to increase their net capacity, at least on a temporary basis, in deferring the planned gradual dismantlement of older but still operative presses.

Conclusion on capacity of German printers for magazine printing

(137) On the basis of the above calculations, the three most important German competitors, namely Schlott, TSB and Burda, would be able to offer additional 197 kt (17 kt spare capacity, 130 kt production shifting, 50 kt net capacity extension) for magazine printing in response to a potential price increase for the printing of German magazines. This theoretical calculation does not, obviously, mean that competitors will make this volume available for the printing of German magazines immediately in response to the creation of the proposed joint venture.

(138) Nevertheless, the calculation shows that a very significant volume of rotogravure capacity could be made available by competitors for the printing of German magazines if the proposed joint venture were to undertake to increase prices for the printing of magazines. The volume which could be made available by German competitors according to these calculations even exceeds the volume of [150-200 kt]* which the joint venture offered on the merchant market in 2003 (and even more the volume of [100-150 kt]*, if the print volume which has become captive for the Parties in the meantime is taken into account).

(139) A price rise would only be profitable for the proposed joint venture if this could increase its profitability on the basis of fewer sales. Considering that, as seen above, the costs of the printing presses (and other machinery) play a very important role in this industry and that rotogravure printers therefore need to achieve as high a capacity utilisation as possible, the proposed joint venture could only forego a limited number of sales if it wished to increase its profitability via a price rise; the number of sales which the Parties could risk to loose would likely be considerably smaller than their entire magazine printing volume. Given the volumes which competitors could make available according to the above calculation, it is unlikely that the Parties could profitably raise prices.

Potential competition of foreign printers

(140) The likelihood of a price increase on the German market for magazine printing is further limited by the presence of several credible potential competitors, in particular RotoSmeets (Netherlands), Quebecor (France), Mondadori (Italy), and to a lesser extent Ringier (Switzerland). In the event of a price rise for the printing of German magazines, the Commission considers it likely that customers could turn to these companies which are currently to be considered potential competitors.

(141) As was seen above, the geographic market for the printing of magazines for German customers is currently limited to Germany. German magazine publishers have so far almost exclusively used German printers, the only exception being one German magazine printed by RotoSmeets.

(142) One of the reasons mentioned in the market investigation for the limitation of the market to Germany were the time-constraints associated with magazine printing and the required proximity of the printer to the area of distribution. However, printers who are fairly close to the German border can be regarded as a potential alternative for the German magazine publishers. The time-constraints associated with magazine printing differ for the various magazines even though they are in general more time-critical than the other printing products. The most time-sensitive magazines are news magazines, economic magazines, TV magazines and people magazines owing to the topicality of their content. A magazine with an exceptionally tight time-schedule is Der Spiegel. Even for this magazine, however, the publishing house confirmed that it would in principle be conceivable to use printers abroad if they are located close to the German border. As the large majority of magazines are less time-sensitive than Der Spiegel, printers close to the German border can be regarded as potential alternatives.

(143) The distance of the sites of the printers mentioned above to the German border does not exceed 350 km, which equals around six hours of truck transportation into Germany; the sites of Roto Smeets and Quebecor (Strasbourg) are even less than 150 km from the German border. The market investigation also confirmed that there are examples that specialised German media forwarders also take up magazines from printers outside Germany. The market investigation further showed that printing outside Germany may lead to slightly higher transport costs (although uniform tariffs for transports inside and outside Germany on the basis of transported tons may also exist). However, the slightly increased transport costs do not inhibit the printing of advertising and catalogues outside Germany and, in the event of a price rise in the printing of German magazines, such disadvantage would no longer be relevant even if it were a factor currently considered by the publishers.

(144) Further reasons for the limitation of the market to Germany are the required know how for the finishing of German magazines and for the preparation of the distribution. German publishers use other methods of finishing and enclosing add-ons, inserts, etc to the magazines than publishers in other countries. Furthermore, different versions of the same edition of a magazine are printed in order to meet the wishes of advertising customers to distribute certain advertisements only on a regional basis (in certain "Nielsen-Gebieten") or not for a specific class of customers (e.g. not to attach advertising post cards or samples to versions distributed to reader circles or to magazines distributed outside Germany). The distribution of different versions to the decentralised system of wholesalers and other points of distribution requires great logistical effort. Printers need to adapt their printing processes to these requirements and finalise the printing of different version of the same magazine so that the media forwarder can take up different versions of the magazine and distribute them in one circuit.

(145) The market investigation showed that German publishers do not consider that printers outside Germany currently possess this know-how (with the exception of RotoSmeets). However, these deficits can be overcome by the acquisition of know-how and, possibly, some investment in finishing machinery. The example of RotoSmeets, which is now able to meet the requirements of the customer for the finishing of a German magazine and for the preparation of the distribution, shows that the necessary know-how can be acquired. It requires close collaboration between the magazine publisher and the printer, and may take some time. In addition to this, competitors located abroad may start to enter the market by using independent German finishing companies in order to enclose add-ons and inserts and prepare the different versions of a magazine for transport and distribution. Since this entails some loss of time, this strategy will mostly be an option for magazines which are comparatively less time sensitive. This solution might further lead to slightly higher transport costs. Again, in the event of a price rise in the printing of German magazines, such a disadvantage would no longer be relevant even if it were a factor currently considered by the publishers. The market investigation revealed that German printers currently use external finishing companies for magazine printing on a regular basis if they are nearby, in particular during peak-times. This shows that the outsourcing of the finishing part to independent finishing undertakings would be an option for printers located outside Germany in order to overcome the deficits in know-how concerning finishing and preparation for distribution and to start entering the German market.

(146) As regards the current price level in the countries neighbouring Germany the market investigation did not show a clear-cut result, owing to the considerable price differences which exist already for individual orders. However, the market investigation indicated that the price level in neighbouring countries tends to be lower or the same as in Germany.

(147) In general, the magazine publishers apparently do not regard the obstacles to foreign competitors as insurmountable. In the market investigation, the majority of magazine publishers pointed, variously, to RotoSmeets, Quebecor, Mondadori and Ringier as viable potential alternatives. They are already repeatedly invited to submit tender bids by German magazine publishers who thereby monitor the market conditions and contact possible alternatives abroad. Each of the four printers has German-speaking account managers, and has acquired experience and reputation in the German market, as all four are currently printing catalogues and/or advertisements for German customers.

(148) RotoSmeets is considered to be the most credible foreign competitor in the German magazine printing market. It already prints one monthly magazine with a print run of [350,000-400,000]* copies and has a German sales office in Bielefeld. Its printing facilities are situated in Deventer and Etten-Leur which are located at 70 and 130 km from the German border, respectively. Taking account of its current spare capacity (on the basis of a maximum capacity utilisation of 95% as applied for the German printers) and its planned capacity extension in the next two years, RotoSmeets could make a considerable additional capacity available for German magazine customers. In addition, RotoSmeets also indicated it could make further capacity available by switching from advertising and catalogue printing. In general, RotoSmeets also would have the incentives to do so, in line with the considerations put forward for German printers above.

(149) Quebecor is considered as a credible potential entrant into the German magazine printing market. It has printing facilities in Corbeil (Paris), Blois, Lille (250 km from the German border), Mary-sur-Marne (330 km from the German border) and Strasbourg with an approximate total capacity of 383 kt according to the data estimated by the Parties. In Strasbourg, Quebecor has a German-speaking desk which currently deals with German catalogue and advertisement customers. Whereas Quebecor's printing facilities in Corbeil, Lille and Blois are currently almost exclusively producing for the French market, the printing facilities in Strasbourg and Mary-sur-Marne show more important rates of export to Germany. As a careful approach, only the Strasbourg and Mary-sur-Marne sites and a maximum capacity utilisation of 95% (as indicated above for the German printers) are taken into consideration. On this basis Quebecor could readily free some capacity for German magazine printing customers.

(150) Mondadori.s printing facilities in Verona and Melzo (Milan) are both located at a distance of 350 km from the German border with a total capacity of [150-200]* kt. Mondadori has extensive experience as a magazine printer in Italy (in-house and for third parties) and could free some capacity for German magazine printing customers on the basis of existing spare capacity and planned capacity extensions in the next two years.

(151) Ringier has a printing site in Zofingen (Switzerland) with a capacity of [25-75]* kt. Data on capacity utilisation has not been provided. Ringier currently prints catalogues and advertisements for German customers.

(152) In the light of the above, at least RotoSmeets, Quebecor and Mondadori are to be considered credible potential competitors on the German magazine rotogravure printing market. On the basis of the figures for 2003, they had at least 32 kt of free capacity which they could dedicate to German magazine publishers. Additional capacity in the amount of around 85 kt could be provided shortly following planned capacity extensions. Further capacity for the printing of German magazines in a volume of more than 50 kt could be made available through shifts in the production mix.

(153) In the light of the above, RotoSmeets, Quebecor and Mondadori can be considered credible potential competitors to which German magazine customers could turn if the joint venture should undertake to raise prices. The calculation for potential competitors of their spare capacity, the planned capacity extensions and the possibility of making capacity available through shifting shows - with all the caveats already set out above for the German competitors - that these potential competitors would be able to make a very significant amount of printing capacity available for the printing of German magazines. This possibility will further reduce the likelihood that the Parties could profitably raise prices.

3.2 Elimination of a Competitor

(154) Further competitive harm, quite apart from considerations that sufficient capacity is available for the printing of German magazines, could theoretically arise from the elimination of a competitor by the concentration. The concentration will remove Springer as an independent competitor. The creation of the joint venture will further lead to the elimination of the currently remaining competition between Bertelsmann/maul-belser and G+J which did not have a coordinated competitive behaviour in the market.

(155) The concentration will thus limit the choice for publishers of magazines, catalogues and advertisements. As most of these publishers organise private tenders prior to the awarding of their printing contracts, in particular those for magazines and catalogues, the number of potential bidders will be reduced. However, it appears from the market investigation that German customers have increasingly invited the abovementioned foreign rotogravure printers to their tendering procedures. Therefore, the effect of the concentration is likely to be alleviated in that respect.

(156) Even if only the German rotogravure printers are considered, customers can still turn to three other significant players Schlott, TSB and Burda with a large installed capacity. In addition, the loss of Springer as an independent competitor may not lead to competition concerns owing to the existence of printers outside Germany, in particular the Dutch printer RotoSmeets, which are considered to be credible potential competitors.

(157) In the market investigation publishers without their own printing facility raised a further question, namely that there could be a problem in their having their magazines printed by a printer which is vertically integrated into publishing. However, the joint venture will not remove an independent printer from the market, but will only bring together printers which were already integrated into publishing. The overall situation for non-vertically integrated publishing houses should therefore not deteriorate in this respect. The market investigation did not indicate any specific segment of magazines in which currently non-vertically integrated printers would face problems on account of a specific competitive situation of their magazines. Furthermore, the market investigation showed that such conflicts of interests are normally solved by contractual safeguards. They might only be critical if a directly competing magazine is printed on the same day. However, the concentration will not lead to a situation in which non-vertically integrated publishing houses no longer have sufficient choice in such circumstances.

3.3 Vertical Integration

(158) Third party competitors raised concerns in the market investigation that the vertically integrated structure of the parties into publishing would create competition concerns. The concerns are based on the consideration that the printing orders placed by the parent publishers constitute the base load for the printing facilities of the proposed joint venture so that the parties only have to fill the remaining printing capacity with orders won in the merchant market. Therefore, third party competitors claimed that the proposed joint venture would enjoy a significant competitive advantage over competitors.

(159) However, such a competitive advantage, if any, would not result from the creation of the proposed joint venture. The parties are already currently vertically integrated into publishing and have the major part of their captive needs for the printing of magazines printed by their own printing facilities. The creation of the proposed joint venture will not lead to any increase of the share of the printing capacity which is used for the parties' captive demand. Contrary to that, the creation of the joint venture is intended by the parties to weaken the direct links between their publishing and printing businesses. [...]*. In any case, the internal print order of the publishing arms will not account for more than [...%]* of the capacity of the proposed joint venture. It will therefore have to fill the remaining capacity with orders won on the market.

(160) Furthermore, the market investigation did not reveal a clear cut picture as to the competitive impact of the vertical integration of the parties into publishing, in particular concerning the "base load" character of the internal print orders. A number of respondents considered that the vertical integration also leads to disadvantages for the printers. The vertical integration makes the printer less flexible as it has to give priority to the internal print orders instead of being able to accept print orders only on the basis of the best commercial results it can obtain on the market.

3.4 Development of overall supply and demand in rotogravure printing

(161) The overall development of supply and demand in the market does not support the expectation of harmful effects resulting from the merger. While overall supply in tendency increases, there are no indications that demand follows suit to the same extent.

(162) As indicated, many competitors plan investments for the near future which will lead to a net increase in capacity in Germany. Apart from this, also the Parties plan to increase their overall capacity by [100-150 kt]* until the year 2008 in Germany. A major investment is moreover currently being undertaken by the Parties in the UK where a new plant is being built in Liverpool with a planned capacity of [150-200 kt]*, around [100-150 kt]* of which are already reserved to a new long-term contract with a British magazine publisher.

(163) This investment will, nevertheless, not be without effect on the German demand and supply situation. It was broadly confirmed by the market investigation, that in the UK, the supply of rotogravure printing is lower than the demand for it. This is one reason for the imports of rotogravure printing services into the UK, mainly from Germany. Some market participants indicated the expectation that some British customers who have imported printing services so far will switch to the new printing facility in Liverpool. This may free some capacity in Germany used before to supply the British customers. Since probably the Parties as well as their competitors will be affected, this may lead to some additional free capacity available for magazine publishers willing to switch to the Parties' competitors should the Parties try to increase prices.

B. Possible Spill-over effects in publishing of Magazines

(164) Pursuant to Article 2(4) of the Merger Regulation, to the extent that the creation of a joint venture constituting a concentration pursuant to Article 3 has as its object or effect the co-ordination of the competitive behaviour of undertakings that remain independent, such co-ordination is to be appraised in accordance with the criteria of Article 81(1) and (3) of the Treaty, with a view to establishing whether or not the operation is compatible with the common market. A restriction of competition under Article 81(1) of the Treaty is established when the co-ordination of the parent companies' competitive behaviour is likely and appreciable and results from the creation of the joint venture.

(165) Both Bertelsmann (mainly through its subsidiary G+J) and Springer are active in the publishing of magazines which are to a large extent printed by means of the rotogravure technique. The publishing of magazines thus constitutes an activity downstream from the joint venture's market for rotogravure printing. As most of Springer's magazines focus on the German market and are sold there, any risk of possible coordination may only occur in Germany.

(166) There is no evidence that the joint venture would have as its object the co-ordination of the Parties' competitive behaviour in the publishing of magazines. Since the printing business may supply an input for the business of magazine publishing, there is, however, the risk that the creation of the joint venture might have the effect tof coordinating the Parties' competitive behaviour in their publishing of magazines.

(167) However, the Commission's investigation has shown that the economic incentives for Bertelsmann and Springer to coordinate their competitive behaviour in the magazine publishing markets are rather limited; therefore, coordination would be unlikely to occur. According to data provided by the Parties the printing costs (excluding paper) account for less than 15% of the total costs of a magazine. The portion of printing costs varies among the different magazines ([...]*) between 5% and 15%. The comparatively limited impact of the printing costs on the price of magazines results in only a minor risk of coordination on the downstream markets for magazine publishing.

(168) The incentives for coordination are further reduced by the relatively low revenues expected in the joint venture's market in comparison with the Parties' revenues from magazine publishing. The Parties' respective turnovers generated by magazine publishing in Germany greatly exceeded their turnover in rotogravure printing: in 2004, Springer's sales of magazines accounted for euro [...]* million in Germany whereas its domestic rotogravure printing activities generated a turnover of euro [...]* million (excluding paper). In the same year, Bertelsmann's (including G+J) sales of magazines reached approximately euro [...]* million in Germany whereas its domestic rotogravure printing activities generated a turnover of euro [...]* million (excluding paper).

(169) In view of the comparatively small impact of the printing costs on the total costs of a magazine and the pre-eminent importance of both Parties' publishing activities as compared to rotogravure printing, the Commission concludes that the creation of the proposed joint venture is unlikely to lead to the coordination of the Parties' competitive behaviour in the downstream markets for magazine publishing.

VII. CONCLUSIONS

(170) For the reasons set out above it must be concluded that the proposed concentration does not significantly impede effective competition in the common market or a substantial part of it, in particular as a result of the creation or strengthening of a dominant position, and that it does not restrict competition within the meaning of Article 2 (4) of the Merger Regulation and Article 81 of the Treaty. The concentration is therefore to be declared compatible with the common market pursuant to Article 8(1) of the Merger Regulation and with the EEA Agreement pursuant to Article 57 thereof.

Has adopted this decision:

Article 1

The notified operation whereby Bertelsmann and Springer acquire joint control of a newly created joint venture within the meaning of Article 3(1)(b) of the Merger Regulation is hereby declared compatible with the common market and the functioning of the EEA Agreement.

Article 2

This Decision is addressed to: Bertelsmann AG

Carl-Bertelsmann-Straße 270

D-33311 Gütersloh

Germany

Axel Springer AG

Axel-Springer-Straße 65

D-10888 Berlin

Germany

Notes:

1 OJ L 24, 29.1.2004, p. 1

2 OJ C ...,...200. , p....

3 OJ C ...,...200. , p....

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

4 OJ C 56, 5.3.2005, p. 24.

5 Turnover calculated in accordance with Article 5(1) of the Merger Regulation and the Commission Notice on the calculation of turnover (OJ C66, 2.3.1998, p. 25).

6 Decision of 15 December 2003 in case COMP/M.3322 - Polestar/Prisa/Inversiones Ibersuizas/JV: OJ C 31, 5.2.2004, p. 19.

7 Conseil de la Concurrence, Avis n° 02-A-01 du 15 février 2002 relatif à l'acquisition de la société European Graphic group ou E2G, filiale de Hachette Filipacchi Presse par la société Imprimerie Quebecor France.

8 A competitor stated that it would be 80% more expensive to run a 32-page offset machine twice than to run a 64-page machine once.

9 Also excluded were all submissions which did not precisely indicate the printing method used or the number of pages.

10 In the UK higher numbers of copies appear to be printed in offset; this appears to be due to the small rotogravure capacity in Britain which does not satisfy local demand. In a presentation by the parties, it was estimated that the UK only had a rotogravure capacity of 343 ktpa whereas the demand in the UK amounts to 673 ktpa. For this reason, publishers resort to offset printing.

11 The parties submitted one example of a magazine with a print-run of approximately 900,000 copies which was printed in offset. However, the Commission's investigation showed that this magazine contains a very high number of inserts and add-ons and an extremely high degree of personalisation. Under these very specific circumstances offset offers a greater flexibility.

12 The parties have submitted the contact details of these Mohn customers very late in the proceedings asking the Commission to proceed cautiously when questioning these customers about their choice of printing method, since there was a high risk of losing these offset-customers to rotogravure printing.

13 Commission Decision of 15 December 2003, case M.3322-Polestar/Prisa/Inversiones Ibersuizas/JV.

14 According to the European Rotogravure Association (ERA), Germany accounts for 46% of the rotogravure capacity installed in the EEA, Italy for 15%, France for 10%, the Netherlands, the UK and Spain for 6-7% each (2003 figures, including data for the new Member States that acceded in May 2004).

15 [...]*

16 Only including TSB.s German printing facility.

17 Only Bauer.s German printing facility.

18 "Others" include Badenia Tiefdruck Offset, Print Forum Druck, Schweiwe Tiefdruck and W. Metz. The figures are based on estimates of the parties; their individual capacity ranges between 14 and 38 kt.

19 The parties estimated the German installed annual capacity to amount to 2608 kt.

20 The figures for the UK do not include the new plants built by Polestar in Sheffield and by Arvato (to be transferred to the proposed joint venture) in Liverpool.

21 Even considering a hypothetical market for catalogue and advertising printing confined to Germany no competition problems would arise. According to the results of the market investigation, the proposed joint venture's market share would amount to [25-30%]* in a hypothetical German market for catalogue and advertising printing: on separate markets the proposed joint venture would have a share of [20-25%]* in a German market for catalogue printing and [30-35%]* in a German market for advertising printing. The proposed joint venture would not be the strongest player in such hypothetical markets; the leader would be Schlott.

22 Case COMP/M.3648 - Gruner + Jahr/MPS.

23 Two printers could not produce any estimate.

24 Even on a hypothetical German market for catalogue and advertising printing the proposed joint venture would not raise competition concerns. Even if competitors shift capacity to magazine printing, this does not mean that the proposed joint venture would be able to exert market power on such a hypothetical German market for advertising and/or catalogue printing.

OPINION

of the ADVISORY COMMITTEE on CONCENTRATIONS given at its 131st meeting on 22 April 2005 concerning a draft decision relating to Case COMP/M.3178-Berteslmann/Springer/JV

1. The Advisory Committee agrees with the Commission that the notified operation constitutes a concentration within the meaning of the Merger Regulation No 139-04 and that it has a Community dimension as defined by that Regulation.

2. The Advisory Committee agrees with the Commission that, for the purposes of product market definition, rotogravure printing of high volume printing orders is distinct from heatset web offset printing. A minority abstains.

3. The Advisory Committee agrees with the Commission that there is a relevant product market for rotogravure printing of magazines. A minority abstains.

4. The Advisory Committee agrees with the Commission that it can be left open whether the printing of catalogues and advertisements in rotogravure is regarded as one single product market.

5. The Advisory Committee agrees with the Commission that the relevant geographic market for rotogravure printing of magazines is limited to Germany.

6. The Advisory Committee agrees with the Commission that the relevant geographic market for rotogravure printing of catalogues can be defined as Germany plus the neighbouring countries (France, Belgium, Netherlands, Luxemburg, Switzerland, Austria, Czech Republic, Poland and Denmark) as well as Italy and Slovakia.

7. The Advisory Committee agrees with the Commission that the relevant geographic market for rotogravure printing of advertisement can be defined as Germany plus the neighbouring countries (France, Belgium, Netherlands, Luxemburg, Switzerland, Austria, Czech Republic, Poland and Denmark) as well as Italy and Slovakia.

8. The Advisory Committee agrees with the Commission that the proposed concentration does not significantly impede effective competition in the common market or a substantial part of it, in particular as a result of the creation or strengthening of a dominant position

a) in the market for rotogravure printing of magazines in Germany.

b) in the market for rotogravure printing of catalogues in Germany plus the neighbouring countries (France, Belgium, Netherlands, Luxemburg, Switzerland, Austria, Czech Republic, Poland and Denmark) as well as Italy and Slovakia.

c) in the market for rotogravure printing of advertisements in Germany plus the neighbouring countries (France, Belgium, Netherlands, Luxemburg, Switzerland, Austria, Czech Republic, Poland and Denmark) as well as Italy and Slovakia. A minority of Member States disagrees with 8 a).

9. The Advisory Committee agrees with the Commission that the proposed concentration does not significantly impede effective competition in the common market or a substantial part of it, in particular as a result of the creation or strengthening of a dominant position

a) in any other relevant geographic market for rotogravure printing of magazines.

b) in any other relevant geographic market for rotogravure printing of catalogues.

c) in any other relevant geographic market for rotogravure printing of advertisements.

10. The Advisory Committee agrees with the Commission that the proposed concentration does not have as an object or effect the coordination of the competitive behaviour of Bertelsmann and Springer in the magazine publishing markets and does therefore not restrict competition within the meaning of Article 2(4) of the Merger Regulation and Article 81 of the Treaty.

11. The Advisory Committee agrees with the Commission that the proposed concentration does not significantly impede effective competition in the common market or a substantial part of it, in particular as a result of the creation or strengthening of a dominant position, within the meaning of Article 2(2) of the Merger Regulation and that the proposed concentration is therefore to be declared compatible with the Common Market and with the EEA Agreement.

12. The Advisory Committee asks the Commission to take into account all the other points raised during the discussion.

<emplacement tableau>

FINAL REPORT OF THE HEARING OFFICER IN CASE COMP /M.3178 - Bertelsmann/Springer

(pursuant to Article 15 of Commission Decision 2001-462-EC, ECSC of 23 May 2001 on the terms of reference of Hearing Officers in certain competition proceedings - OJ L 162, 19.6.2001)

On 4 November 2004 the Commission received notification of a proposed merger by which the undertakings Bertelsmann AG (Germany), its solely controlled subsidiary Gruner+Jahr AG & Co. KG (Germany), and Axel Springer AG (Germany), acquire joint control of the undertaking NewCo by way of purchase of shares in a newly created company constituting a joint venture.

Having examined the information submitted by the parties to the proposed merger and conducted a market survey, the Commission concluded that the merger raised serious doubts as to compatibility with the common market and the EEA Agreement. On 23 December 2004, therefore, the Commission initiated the procedure provided for by Article 6(1)(c) of the Merger Regulation.

Following a detailed market investigation, the Commission concluded that the proposed concentration did not significantly impede effective competition in the common market or a substantial part of it, in particular as a result of the creation or strengthening of a dominant position, and that it did not restrict competition within the meaning of Article 2 (4) of the Merger Regulation and Article 81 of the Treaty. Accordingly, no statement of objections was sent to the parties. In the course of the market investigation, the parties were granted access to key documents under section 7.2 of DG Competition Best Practices on the conduct of merger control proceedings.

The case does not call for any particular comments as regards the right to be heard.

Brussels, 27 April 2005.

(signed) Karen WILLIAMS