CCE, 2 mars 2001, n° M.2256

Dear Sir/Madam,

Subject: Case No COMP/M.2256-Philips/Agilent

Notification of 1 February 2001 pursuant to Article 4 of Council Regulation No 4064-89

1. On 1 February 2001 [1], the Commission received a notification of a proposed concentration pursuant to Article 4 of Council Regulation (EEC) No 4064-89 [2] by which the Dutch undertaking Koninklijke Philips Electronics N.V. ("Philips") acquires within the meaning of Article 3(1)(b) of the Council Regulation control of the medical business (Healthcare Solutions Group, "HSG") of the US undertaking Agilent Technologies Inc. ("Agilent") by way of purchase of assets.

[1] The notification originally submitted on 22 December 2000 was declared incomplete on 23 January 2001. The parties submitted the missing information on 1 February 2001.

[2] OJ L 395, 30.12.1989 p. 1; corrigendum OJ L 257 of 21.9.1990, p. 13; Regulation as last amended by Regulation (EC) No 1310-97 (OJ L 180, 9. 7. 1997, p. 1, corrigendum OJ L 40, 13.2.1998, p. 17).

2. After examination of the notification, the Commission has concluded that the notified operation falls within the scope of Council Regulation No 4064-89 and does not raise serious doubts as to its compatibility with the common market and with the functioning of the EEA Agreement.

I. THE PARTIES

3. Philips is a multinational company active in the manufacture and sale of electronic products for domestic appliances and medical purposes. In the health care sector, Philips manufactures, in particular, medical imaging equipment (x-ray, computer tomography ("CT"), magnetic resonance ("MRI"), nuclear medicine and ultrasound).

4. Agilent [3], headquartered in the United States, operates in the research, development, manufacture, marketing and sale of communication, electronics, life science and healthcare products. Its medical care business branch HSG produces mainly ultrasound imaging equipment, patient monitoring devices and cardiac therapeutic equipment (such as resuscitation and ECG).

[3] The company was spun off from the Hewlett-Packard Company ("HP") in 1999 as part of HP's strategy to focus on its core competencies in computers and printers.

II. THE OPERATION

5. Philips and Agilent entered into an Asset Purchase Agreement (APA) on 17 November 2000. The APA provides for the purchase by Philips of all of the assets of HSG. As a result of the transaction Philips will thus exercise sole control over HSG. The proposed operation constitutes a concentration according to Article 3 (1) (b) of the Merger Regulation.

III. COMMUNITY DIMENSION

6. The undertakings concerned have a combined aggregate world-wide turnover of more than EUR 5 billion [4] (Philips: EUR million 31,459; HSG: EUR million 1,340 million). Each of the two undertakings concerned have a Community-wide turnover in excess of EUR 250 million (Philips: EUR billion 13,11; HSG: EUR million 265), but they do not 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. It does not constitute a co-operation case under the EEA Agreement pursuant to Article 57 of that Agreement.

[4] 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, p25). To the extent that figures include turnover for the period before 1.1.1999, they are calculated on the basis of average ECU exchange rates and translated into EUR on a one-for-one basis.

IV. COMPATIBILITY WITH THE COMMON MARKET

A. Relevant product markets

7. The only areas where the activities of Philips and HSG overlap is the production and sale of ultrasound imaging equipment. HSG specialises in the manufacture of ultrasound machines, in particular for cardiac applications. Philips entered the ultrasound segment in 1998 through the acquisition of ATL, a US based company active in diagnostic ultrasound systems for diverse applications including radiology, cardiology, obstetrics and gynaecology.

8. Ultrasound machines are inter alia used for medical diagnostic imaging purposes. Medical diagnostic imaging can be defined as the capture of data for the production of images of internal organs of the human body by non invasive means for the purpose of clinical diagnosis (see for instance case COMP/M. 1298-Kodak/Imation). Other devices used for medical imaging purposes are X-ray, MR (magnetic resonance), CT (computed tomography) or NM (nuclear medicine) [5]. The parties argue that, although these (other) medical imaging products rely on different technologies, they exercise a competitive restraint on ultrasound machines since similar results can be achieved with these techniques and production costs of more expensive imaging products (like CT or MR) are in decline. There are however, a number of important differences between ultrasound machines and other imaging products relating to unit production costs, price per application and the results which can be achieved on the patient. According to customers, there are very few instances where ultrasound competes with other imaging techniques for a given application, and even in these cases, relative prices are not normally the main parameter of choice. It can therefore be concluded that ultrasound equipment constitutes a separate product market from other imaging devices. The exact definition of the latter does not need to be discussed further since there are no overlaps in imaging devices other than ultrasound machines.

[5] In contrast to X-ray, CR, NM and ultrasound are generating digital images by non- conventional means.

9. The parties are of the opinion that the relevant product market is the overall market for ultrasound machines including all types of clinical applications. The parties argue that there is growing demand in Europe for multipurpose (or multi-speciality) machines, which can be used across various applications and shared between various hospital departments. The parties also stress that the market is moving towards scalable systems, that is, systems using standard PC chips and NT software allowing to use the same platform across different medical applications. These systems are deemed to greatly facilitate the use of multipurpose machines for the customer. However, the parties themselves state in the notification that most of today's ultrasound machines have dedicated chips and software. Scalable machines are at present being developed by a number of manufacturers. They aim predominantly at the lower end of the market. It is at present unclear whether and to what extent scalable technology may in the future have an effect on the overall market dynamics.

10. The market investigation has revealed that most competitors distinguish between different ultrasound applications, such as radiology, obstetrics, gynaecology and cardiology [6]. As pointed out by the parties, cardiology applications are covered not only by dedicated machines but also multipurpose machines, which can also be used in other clinical areas (such as obstectrics, radiology, etc.). This does, however, not indicate the definition of an overall market for ultrasound products since ultrasound machines suitable for the examination of the heart have to meet different technical requirements than other ultrasound machines. Ultrasound machines used for cardiac applications (whether they are dedicated or multipurpose) need to generate images of a moving organ which is located under the ribs. Therefore, special imaging techniques, software and particular transducers are needed for cardiac ultrasound, which are not needed for other applications. [7] In addition, manufacturers tend to have separate sales and marketing forces for cardiac ultrasound. Furthermore, cardiac ultrasound is the application where most technological progress and innovations have occurred in the past 5 years. Cardiology is therefore analysed as a separate segment in most of the clinical research publications. For these reasons it cannot be ruled out that ultrasound machines used in cardiology constitute a product market distinct from machines which do not provide for this application.

[6] Those are the ones most frequently mentioned by independent research institutes.

[7] Cardiology almost exclusively encompasses the technique known as "echocardiography", a method where ultrasonic waves directed through the heart are reflected backward or echoed when passing from one type of tissue to another.

11. In addition, some competitors distinguish between low-end ultrasound machines (< 40,000 USD), mid-range products (between 40,000 and 100,000 USD) and high-end machines (>100,000 USD). This segmentation is also reflected in some of the medical publications, such as the Clinica Report: "The World Markets" (1.6. 1998), which identifies the following ranges: standard ultrasound products (below 50,000 USD), mid and high-end ultrasound products (between 50,000 and 90,000 USD) and premium level products (equalling or above 150,000 USD).

12. The parties reject the need for a segmentation according to price ranges, arguing that cardiology machines within one range can compete with products located in neighbouring ranges. The parties submit, for example, that Agilent's Sonos 5500 or Sonos 4500, which are both high-end machines, compete with Acuson's Aspen or GE's Logis, which are both mid- range products. The parties also argue that there is a high degree of supply-side substitutability. Finally, the parties maintain that high-end systems have in recent years lost market share because, as a result of generally improved technology, a growing number of examinations can be performed by medium-range machines. In this view, the distinction between the different performance levels has become increasingly blurred. By contrast, the parties themselves explain in the notification (confirmed by the market investigation) that the ultrasound market is a technology and R&D driven market where technological developments largely determine competitive positions and where technological innovations continue to be introduced primarily at the top-end of the market. In any event, it can be concluded that there are a number of new technologies at the development stage (such as real-time 3-D imaging, ultrasound contrast agents, information management, etc.), which will be introduced first for products in the premium and high-end segments although at a later stage these new technologies can become standard for use also in mid-range equipment.

13. The results of the market investigation do therefore to a certain extent support a distinction between different product ranges. High-end machines demand periodical software up-dates and are focussed on specific customer groups. These machines typically provide a number of automated calculation tools and a higher level of contrast resolution than mid- range and low-end products and are in particular purchased by university hospitals and large clinics engaged in research activities. Certain mid-range and most low-end machines are sold to smaller clinics or private doctors. These devices are in general used to identify the existence of a health problem relating to the heart while more sophisticated machines are needed to exactly locate the source of the disease. [8] The market test showed that prices and performance of the different levels (low/mid/top-end products) are different in a material respect, e.g. low-end machines have basic imaging capabilities but limited contrast capabilities. Differences also exist with regard to the capacity of the machines. With low-end machines 10-15 patients can be examined per day, whereas mid-range devices allow a number of 20-25 patients a day. For these reasons it cannot be ruled out that the cardiac ultrasound market can be further subdivided into different product ranges.

[8] Basic systems offer two-dimensional black and white imaging. More advanced systems offer the so-called "Doppler" technique, which allows blood flow patterns and velocity to be observed. The most expensive systems offer colour in conjunction with "Doppler" (direction and velocity of blood flow is characterised by colour and intensity).

14. In the present case, however, it is not necessary to decide on the exact product market definition since with every possible alternative market definition competition concerns do not arise from the proposed transaction.

B. Relevant geographic markets

15. The parties argue that a European market exists for ultrasound machines overall as well as for the separate segments outlined above. Both Philips/ATL and Agilent manufacture their products in various parts of the world and distribute their ultrasound equipment world-wide. Transportation costs are not relevant since ultrasound-machines are high value products. The parties state that public hospitals are to a large extent subject to EEA-wide public tendering procedures. They add that in those cases where EC public procurement rules do not apply, public hospitals would often apply similar tendering procedures all over Europe.

16. The Commission in the case No. COMP/M. 1298-Kodak-Imation left open the question whether competition for diagnostic imaging products takes place on a European or national basis. In the present case, the market investigation has confirmed some of the parties' arguments. On the other hand, several competitors and customers also stated that demand characteristics are different between EEA Member States and that procurement procedures of public administrations were also different. Some competitors also mentioned that different reimbursement schemes had historically led to significantly different competitive product profiles and pricing across Europe. In addition, significant service and maintenance requirements, which are partly performed by the manufacturer, limit customers' ability to bypass local sales organisations. Suppliers may, thus, be able to price discriminate between customers in different geographic locations.

17. In the present case, however, the ultimate geographic market definition can be left open since competition concerns do not arise from the operation on any national market or at a European level.

C. Assessment

All ultrasound

Impact of the concentration

18. Both Philips/ATL and Agilent sell a variety of ultrasound machines, which can be specifically configured for cardiac diagnosis: Philips/ATL offers the HDI 5000, the HDI 5000 CV [9], the HDI 5000 SonoCT, the HDI 3500, the HDI 3500 CV and the HDI 1500. All these machines are multipurpose machines. Agilent's machines are the Sonos 5500, the Sonos 4500 (dedicated machines) and the ImagePointHx (multipurpose).

[9] CV: cardio-vascular applications only. These are dedicated machines.

19. The total market for all ultrasound equipment in 1999 in the EEA is estimated by the US marketing research institute "Frost & Sullivan" at 924.8 million USD. The market shares of Philips/ATL and HSG as well as of their main competitors at a national and EEA level are set out in the table below:

All ultrasound

>TABLE POSITION>

20. The parties have additionally submitted up-dated market information for 2000 based on their own estimates, according to which their combined national and EEA-wide market shares would be slightly lower ([15-20%] at EEA level). The parties submit that they have been losing market shares since 1998 ([between. 5 and 10% for ATL and for HSG at EEA-level]) and that their main competitor, GE, has gained more than ([15-25%] in 1999 and 2000.

21. At EEA level, the parties after the concentration would be approximately as large as Siemens/Acuson with around 22%, GE being the third largest competitor with around 13% market share. There are also three smaller players having a share of approximately 9-10% each.

22. The parties after the concentration will be market leaders in France, Spain, the UK and the Benelux but will be challenged by a number of major competitors (in particular GE and Siemens/Acuson) who have comparable market shares. The combined market share of ATL and HSG would exceed 25% only in the Benelux area (27.4%). Again, there are other strong competitors active on the Benelux market such as GE or Siemens/Acuson. [10]

[10] The parties argue that Belgium, Luxembourg and the Netherlands (as well as Greece, Austria and Portugal and the Scandinavian countries) can be looked at as relevant regional markets. The parties argue that sales volumes in these countries are so small that they are unreliable. Furthermore, the parties do not have their own sales forces in those countries. In any event, they submitted credible information according to which market shares at a national basis would not deviate to an appreciable extent from regional market shares if taken as an average over 4-5 years.

Conclusion

23. It can therefore be concluded that the concentration will not create a dominant position of Philips/HSG in the overall ultrasound market.

Cardiology applications (multipurpose and dedicated machines)

Impact of the concentration

24. The total market for cardiac ultrasound in 1999 in the EEA is estimated by Frost & Sullivan at 218.6 million USD. The market shares of Philips/ATL and HSG as well as of their main competitors at a national and EEA level are set out in the table below:

Cardiac ultrasound

>TABLE POSITION>

25. The parties again submitted up-dated market information for 2000 based on their own estimates, according to which their combined national and EEA-wide market share would be slightly lower.

26. Two competitors estimate the parties' combined market share in cardiac ultrasound to be higher (up to 50% in certain Member States and at EEA level). One medical research institutes also reports somewhat higher market share figures than given by the parties (Diagnostic Imaging 1999 Market Report [11]). According to this source HSG's (then HP) and ATL's combined market share would be around 45%. The Market Line Study 1997 partly reports significant market shares for ATL and HSG in certain EEA-Member States (over 50% in Italy and France, 45-40% in NL, Germany and Denmark). Several reports refer to HSG as the leading player in cardiovascular ultrasound.

[11] Europe 99: Market trends

27. The parties also supplied market share estimates for the different ranges (high-end, medium-end and low-end). In the low-level segment HSG is hardly present ([0-5%] at EEA level) and therefore there is hardly any overlap (ATL has [between 5 and 13%] at a EEA level and its market shares do not exceed 20% in any national market). Combined market shares of Philips/ATL and HSG for medium-end machines are below [between 35 and 45%] both at national and at EEA level. In the medium-range segment, not only GE and Siemens/Acuson but also a number of smaller manufacturers compete against Philips/ATL and HSG, such as Esaote, Aloka, Toshiba, Kretztechnik (Medison) or Hitachi.

28. HSG is strongest in the high-end segment ([between 25% and 35%] at EEA level and market shares exceeding [35-45%] in a number of national markets). Philips/ATL has a European market share of [between 5 and 15%] and market shares of [25-35%] in France; otherwise its market share does not exceed 10% in any other Member State. The parties' combined shares would thus be above [35-45%] in UK [ ], the Benelux [ ] and Italy [ ]. At a European level, the combined market share of Philips/ATL and HSG for high-end ultrasound machines would be around 40%.

29. After the operation there will still be three to four major suppliers of high-end cardiac ultrasound products (Philips/HSG, GE, Siemens/Acuson, and, in some areas, Toshiba). Each of these companies holds leading market positions in a number of product markets (e.g. MR, X-Ray, CT etc.) and covers a wide range of medical equipment, enabling it to provide hospitals with integrated product packages across a range of clinical departments.Toshiba appears to have in recent years lost ground to the three market leaders and most customers do no longer consider it as strong as the other three firms. The remaining players (the Italian company Esaote, the Japanese companies Hitachi and Aloka and the Korean manufacturer Medison) are comparatively smaller firms with narrower product ranges. Customers at present consider them as credible competitors only in the lower end of the market and certain niche products (e.g. endoscopy). Aloka, nevertheless, has a significant market share (approximately 10% at EEA level) in the cardiac ultrasound market.

30. However, despite the parties' important position after the concentration in terms of market share, customer base and related know-how the market test showed that Philips/ATL and HSG will not obtain a dominant position after the concentration in cardiac ultrasound or any particular segment thereof. The main conclusions of the market test are outlined below.

Market positions are largely determined by innovation and do not remain stable over time

31. As explained in the notification and confirmed by the market investigation, the cardiac ultrasound market is R&D intensive and largely driven by technological innovations which take place at relatively rapid pace, on average every 4-5 years. Demand for ultrasound is growing faster than for other diagnostic imaging devices (5-7% per annum). The rapid innovation rate of ultrasound allows competitors, who manage to place a new product on the market, to gain market shares relatively quickly, while established products might lose out. HSG (then HP, Hewlett-Packard) prior to 1998 was by far the market leader in Europe in cardiac ultrasound. Later on, GE and Acuson managed to improve their position significantly. GE bought Diasonic/Vingmed, a Norwegian company, which had made considerable investments in developing a high-end CV ultrasound machine ("Vivid"-Line), while Siemens (who had not been active in cardiac ultrasound before) acquired know-how from the US company Acuson (which it acquired in 1998). These developments have resulted in a significant price decrease for top-end equipment but also in a change of market shares of the main competitors (HP lost ground on GE and Siemens/Acuson).

32. The most important recent innovations include computed sonography (developed by Acuson) in the early 80s, the Doppler Colour Flow (developed by the Japanese company Aloka and HP) in the mid-80s and digital ultrasound (developed by ATL) in the early 90s. New developments in manufacturers' research pipeline include real-time 4-D imaging, ultrasound contrast agents and the development of scalable platforms (see above). Two smaller manufacturers have stated that they intend to release new products aimed at the high- end market based on such technological innovations. It can therefore be expected that market positions of the main suppliers' products will continue to be challenged by new innovative competitors.

The concentration will not remove the closest substitute

33. The market investigation has concluded that HSG's and ATL's products in cardiac ultrasound are not the closest substitutes. Competitors in the majority of cases considered GE's "Vivid V" and Siemens' "Sequoia" product lines as being the closest substitutes for Agilent's high-end cardiac ultrasound devices, the Sonos 5500 and Sonos 4500, as well as for ATL's HDI 5000 (CV) and HDI 3500 (CV).

34. This finding was confirmed by an economic study supplied by the parties. The analysis was based on "win/loss"- data recording the results of tenders won and lost by HSG in cardiac ultrasound between 1998-99 and 2000 and concluded that for []% of all projects won by HSG, Siemens/Acuson was second placed whereas for [ ]% of projects won by HSG, GE was ranked second. In [ ]% of all cases ATL was second placed. On the other hand the study revealed that [ ]% of all projects lost by HSG between 1998-99 and 2000 were won by GE, [ ]% by Siemens/Acuson and around [ ]% by ATL.

35. The conclusion which can be drawn from the study is that for HSG's cardiac ultrasound machines (both for the Sonos 5500 and the Sonos 4500) GE and Siemens/Acuson are the strongest challengers on both projects won and lost. ATL is generally the third ranked. Therefore, it can be assumed that Philips/HSG would not be in a position to increase prices for one or both products without facing competitive constraints by the other first-tier suppliers.

The market is not capacity-constrained

36. For high-end ultrasound machines there are at present no capacity constraints in the market since the major players Acuson and GE and, to a certain extent, also the multinational company Toshiba have the necessary resources to satisfy additional demand in case of a unilateral price increase. All of these players (Toshiba with some exceptions) supply a full range of cardiac ultrasound products and have the necessary know-how and distribution infrastructure to serve customers' needs. Imports are generally not restricted. At least GE and Siemens/Acuson also have a comparable degree of brand awareness with their customers.

37. For mid-range and low-end products there are no capacity restraints either since not only the first-tier manufacturers are active in these segments but also a number of smaller suppliers, such as for example the Italian company Esaote (who recently acquired the US ultrasound company Pie Medical from Philips) with its EU 5 in medium-range cardiac ultrasound, Aloka (market-leading Japanese manufacturer, benefits from R&D carried out from its parent company) or Medison (Korean firm, acquired a significant foot-hold in Europe through the acquisition of the Austrian ultrasound company Kretztechnik in 1996).

Customers do not in general face disproportionate switching costs

38. Customers of cardiac ultrasound machines include on the one hand large hospitals and universities (mainly top-end and mid-end products) and smaller clinics and private doctors on the other hand (predominantly mid-range and low-end products). As a rule customers throughout the EEA purchase the majority of their diagnostic imaging equipment via public or private tenders (less than 200,000 EUR). Usually there are separate tenders for ultrasound equipment and other imaging devices. Public tenders are subject to the EC public procurement directives. Customers report that basically all suppliers active in a particular product segment usually participate in public tenders.

39. Customers do not encounter significant obstacles when changing suppliers of cardiac ultrasound machines, neither in technical nor in financial terms. Most customers require their suppliers' products to meet DICOM ("Digital Imaging Communication") standards, which guarantee interconnectivity to the effect that data on patients can be exchanged between machines of different manufacturers. Therefore, ultrasound equipment from different suppliers can be used without major difficulties in one and the same hospital environment. When switching to different suppliers' products, the hospital personnel and doctors concerned have to be trained on the new equipment. However, most suppliers offer training as a part of their pre- and after sales services. Several suppliers also offer service for other than their own ultrasound products.

Market entry by second tier suppliers can be expected if prices were to rise

40. According to certain sources (see for instance Clinica Report 1998) margins in the high- end segment appear to be attractive. The market investigation showed that several smaller manufacturers, who are at present primarily active in the medium and low ranges are at present preparing to enter the segment for high-end ultrasound machines. If prices for high- end ultrasound machines were to increase entry from these second-tiers can reasonably be expected to constrain the parties' pricing behaviour. Two other companies, who are already credible competitors for mid-range products also aim to enter the high-end cardiology segment.

V. ANCILLARY RESTRAINTS

41. Agilent and Philips have agreed upon a number of restrictions following the present transaction. Under Article 6.10 of the APA Agilent agrees that for a period of five years, Agilent and its subsidiaries will not engage in any competing activity, that is, in the manufacture and sale of medical products to third parties. This clause can be regarded as being ancillary to the present transaction to the extent that it covers the products which are at present manufactured and sold by HSG and the countries where these products are currently sold to. In the absence of any exceptional circumstances indicated by the parties the non- compete clause can be regarded as ancillary to the present concentration for a period of only three years from the proposed transaction. The non-compete clause, however, cannot be considered ancillary in as far as it concerns purely financial interests of Agilent in competing businesses, even if these interests exceed 10% (see for example Case Comp/M.301- Tesco/Catteau, par. 14).

42. Second, under the terms of Article 6.11 of the APA Agilent agrees that for a period of maximum 18 months it shall not solicit or hire any key personnel employed by Philips. Philips also agrees that for 18 months it shall not hire key personnel employed in the business retained by Agilent. While the first obligation can be regarded as ancillary to the present operation as far as it relates to key personnel employed in the healthcare division or related functions, the second cannot since it involves obligations for the buyer and relates to businesses not concerned by the operation.

43. Finally, Agilent agrees under Articles 6.8 (a) and 7.2 (d) of the APA to continue to provide to Philips certain transition services for a period of 12 months, which the parties deem necessary for the continuation of the transferred business by Philips. These services relate inter alia to information technology, support of existing customers, service parts and supplies and the collection of receivables. These supply and service obligations can be regarded as ancillary to the operation and as proportionate to enable the transfer of the assets to Philips under reasonable conditions.

VI. CONCLUSION

44. For the above reasons, the Commission has decided not to oppose the notified operation and to declare it compatible with the common market and with the EEA Agreement. This decision is adopted in application of Article 6(1)(b) of Council Regulation (EEC) No 4064- 89.